Boris Mann's Blog - Future of web, mobile, and retail

Vancouver's talent are like raw resources

2013-03-17T17:30:00-07:00

I had a brief clip air on Global TV BC this evening, commenting on the news that Facebook is opening a temporary office in Vancouver. Here’s a link to the written article. Thanks to Greg for tracking down the Global TV News Hour clip (starts at 15:20).

I do think that the difficulty in getting a US Visa is a contributing factor to make Canada / Vancouver an attractive place to put an office. It was back in July 2007 that the Microsoft opening an office news broke. Looking back, the Microsoft office out in Richmond was basically a non-impact on the local community. So, where Facebook puts its office and how much it interacts with the local community will be the determining factor on the potential impact of having them here.

I’d like to talk about local Vancouver talent. My general feeling is that the developer and designer talent that Vancouver does have are treated like raw resources. Like our logs and other natural resources, we do very little “secondary processing”, and the best are shipped off elsewhere.

Igor Faletski, CEO & Co-founder of Mobify, did a great write up right after the news broke: What the new Facebook office means for the Vancouver tech scene

I agree with his summary, that this is a good thing - Facebook will bring talent from around the world to Vancouver (and some will stay), and that another strong link between here and the Valley is a good thing for the community.

The main downside is the hiring pressure. Mobify is a growing company, and in his last blog post covering what startups should focus on, Igor said “Companies - and people - that don’t master hiring can’t scale, so spend at least a quarter of your time on it.”

My observation on “talent in Vancouver” has always been three fold:

There are few companies for top / senior talent to work at in Vancouver. This is changing (HootSuite, Mobify – and as of recently, Salesforce and Amazon)
People are underpaid here compared to the rest of Canada, and really underpaid compared to anywhere in the US. More local competition will raise the bar here.
We have lots of great juniors who come from our local universities and other institutions. Many get better paying jobs elsewhere (#2), or leave once they’re intermediate since there is limited room for career growth (see #1).

Also, with generally smaller companies and/or “branch office” locations (and so nowhere for people in the following disciplines to gain experience), Vancouver is in dire need of soft skills in marketing, business development, sales, product management, recruiting, etc. etc. etc.

Since these are the critical factors in scaling a company past a couple of co-founders, this is likely an even bigger issue than “I can’t find a Ruby programmer to hire”.

Also related to being underpaid, this has had an interesting side effect - since salaries are low/career opportunities are few, doing a startup feels less risky. However, since we also have very limited angel capital, the best startup teams again head south where there is more angel capital at higher valuations.

Is there lots of “great” talent in Vancouver already? I’d say that there are many talented juniors, but in general it’s hard to find people who are more senior / have lots of relevant experience. However, because of the generally low pay, hiring people away from existing companies (that likely also have less interesting career paths available) is going to be easy for new entrants like Facebook, Salesforce, and Amazon.

Comments on Twitter (see full collected responses on Storify) underline this:

Dan Udey: “after I was hired, I watched the founders interview for six months without hiring”

Dan works at A Thinking Ape (aka ATA), which is a company that relocated to Vancouver from Silicon Valley. ATA has done a great job of running new grad focused job fairs across Canada. And that’s the point (again): hiring great people is hard, and we need to a) invest in getting good at it and b) great people are everywhere, and we have to go looking for them.

We need to invest in our talent to keep them here with both competitive salaries and great career opportunities. Or they will leave. Just like the raw logs that we export.

The Flickr Question

2012-11-24T11:54:00-08:00

It’s that time of year again, when I face the same question: my Flickr Pro subscription has expired, and I need to pay so that all my photos are accessible online. But maybe I should self host? Is paying for a Flickr Pro subscription still worth it?

Don’t get me wrong - I think the “Flickr deal” is still fantastic: about $2 per month for unlimited storage of all my photos. From a pure storage perspective, even Amazon S3 would cost me more than that.

Flickr was one of my first loves when it came to online community. When I could finally give the company money, it felt that much better: I was supporting something great by paying for it.

On top of just photo storage, I get a great API, thumbnail generation, maps, tagging, and so on and so on.

Community? Not so much, anymore, although perhaps that’s also a function of me taking less photos and more iPhone snaps. I do miss the great community that built up there, but I’m used to the ebb and flow of taste and online fashion. Heck, I’m even posting to G+ these days!

What options do I have outside of Flickr?

OpenPhoto

OpenPhoto looks promising – and in fact, they overheard a discussion on Twitter between myself and Darren Barefoot¹.

OpenPhoto is both a desire by people to self host (own their data), to use their own domains (own their permalinks), and in general not to be beholden to advertising or commercial entities.

But it has a couple of issues. The first is the permalink issue (here’s OpenPhoto’s blog post on the subject). All of that is great - I would much rather have my images available at photos.bmann.ca/photo/1234 than on Flickr or Facebook.

But what a LOT of people love Flickr for, is using it as the place where they upload source images, and then they embed the Flickr thumbnails in their blog. And that’s where the OpenPhoto “bug” with their permalinks story is: the thumbnails they generate (and it looks like they only generate one size right now?) are hosted on a separate server with an address that is not your own.

My test site is at boris.openphoto.me. The original image embedded below is here, but the thumbnail is on the awesomeness subdomain.

Of course, OpenPhoto is committed to open-ness and hosting your own data, so I assume they’ll fix this at some point. This GitHub issue thread on writing metadata (favorites, comments, etc.) directly into the photo file itself is a great example of some forward thinking around this.

Self Hosting

The other issue I have with OpenPhoto is their approach to even writing the software. Or rather, their assumptions.

It used to be that self hosting meant running a Linux box / LAMP stack off behind your own cable or DSL connection. Perhaps a cheap managed server or (more common these days) a VPS of some kind.

But I don’t think that’s the path forward. The expertise needed to manage the entire stack of an operating system, a web server, a database server, AND the programming language you’ve chosen is too much complexity.

As I wrote in The New Hack Stack, I see the way forward being PaaS: choose a programming language, choose a PaaS. The database should be fully managed as well.

In short: I don’t want to take a step backward and have to run a full server stack just to have to run OpenPhoto. I’d love to see an alternate approach from the team where a git clone, some basic config file editing, followed by a push to Heroku / Staccato / PaaS of your choice (which probably means ditching PHP - I had to get a language dig in here somewhere!).

Heck, maybe a central photo management app that keeps your account, settings, lets you uploads centrally, but publishes the photos / thumbnails / etc. to relatively flat files on your own storage space on S3, Dropbox, etc. is an interesting way to make this even easier.

What’s the answer to the question?

For now, I’ll keep paying for Flickr Pro.

It’s cheaper than any other storage option
No one else has solved the permalink issue
I want great user-supported web services to exist

I am not worried about my photos or the portability of the data there: Flickr was one of the very first of the new breed of websites to have a really great API, and there are tons of end user friendly export systems.

I am worried that Flickr will continue to slide along with the rest of Yahoo, but with Marissa Mayer at the helm, all sorts of changes are coming, that on balance look to be good.

I’ll keep experimenting with OpenPhoto and other solutions. I really like having my main site on Amazon S3 and this blog on Heroku, and paying for the storage on my own.

You can find me and my photos on Flickr as boris, and on Instagram as bmann.

Footnotes

¹: Here’s some of the Twitter conversation embedded via Storify:

Working with Contractually

2012-09-04T11:54:00-07:00

I’ve been working with Contractually for the past 2 weeks¹. I’ve always loved Martin’s vision for Contractually² – both the concept of moving past “digital representations of paper” to more fluid, data-native documents, as well as modernization of the legal business.

Broadly speaking, I’m doing Business Development. I’ve been trying the role title “growth hacker” on for size, but the phrase is a bit problematic. It has gone from a useful description for a role that understands digital tools & metrics and pairs them with more traditional marketing and business development straight to a buzz word akin to “social media guru”. That is, if someone is using the label “growth hacker”, there is a good chance that they don’t know what the hell they are talking about.

Still, I find the term interesting. I’ve spent a lot of time talking to Ian Walker over at Perch about a variety of “glue” roles within growing web startups. We put on a Business Tools Hacking event that had about 20 people attend and share both what tools they used and what process they were supporting. More on this in a wrap up post about the event.

In any case, some of the first results of my work are now visible as the new Contractually marketing site is up. We ended up selecting Unbounce³ to run the handful of front page landing pages that describe the product and lead people to the sign up page. This is by no means finished, but we now have a framework for testing various hypotheses about different markets and channels. Stay tuned for further updates and new Contractually features.

Connection Karma

What else have I been doing? I spent a good chunk of time taking some time off with some trips down to Seattle, figuring out what I wanted to be spending my time on, and an assortment of startup meetings & events. For me, meetings, introductions, and seeing what happens when you spend time getting to know people fall under the broad heading of Connection Karma.

I’ve decided to focus on consulting contracts like the one with Contractually, as well as smaller projects that focus on strategic advice, marketing & business development projects, and in general helping companies to build learning organizations. What does that mean? Culture, tools, testing & experimentation across all parts of an organization – as well as the processes to gather everything together and make that information available to the people that need it⁴.

This is actually the first time in 8 years that I’ve had time in my schedule for projects of opportunity. I can’t shake the drive to be involved in helping to build the startup ecosystem here in my home town of Vancouver, so you’ll see some events & projects related to that. I’ve got some ideas of my own, but they’re (purposefully) not big ideas, more like an excuse to work with people & technologies that I enjoy working with.

I’m talking to a handful of other organizations and people about working together. For now, you can find out a bit more about working with me at my Connection Karma site.

Back to Contractually

Contractually is a great tool for anyone that needs to deal with contracts – freelancers that either don’t use contracts as often as they should (or at all!), companies that find the dance of emailing Word documents back and forth a pain, or even your HR department, that needs to keep track of employment agreements, IP assignment, and the handful of other documents that every employee needs to sign. Simplify your contracts process. Create, negotiate and e-sign your contracts online.

If you’re a web developer, a growing mid-size company, a photographer, or anyone else that works with contracts, I’d love to talk to you about how you use contracts today.

Footnotes

¹ I’ve been telling people that I left iQmetrix back at the beginning of July in person, but haven’t made a big announcement about it. Although if you watch my LinkedIn or Twitter profiles, I did flip the switch and change my profile descriptions back then.

² Martin & Contractually were part of the first cohort of startups at Bootup Labs.

³ I love both Unbounce the tool, and Unbounce the local Vancouver startup team. Designed for landing pages, I’m increasingly seeing it as a great tool to match with one-off domains, events, and even creating the 4 - 6 pages that make up the front end marketing pages of many web apps.

⁴ I’ll do everything I can to avoid uttering the phrase “knowledge management”, since I fundamentally don’t believe that knowledge can be managed. More recently, we’ve seen the emergence of streams/flows, which more clearly shows that knowledge is not something that is locked away somewhere. At the same time, it feels like lots of organizations need to learn that email is where information goes to die the hard way.

StartupRiot Seattle 2012

2012-08-11T12:54:00-07:00

I’m just getting back from 2 days in Seattle. In general, I’m trying to spend more time there, getting to know what’s happening in the Seattle community and seeing if Vancouver can connect more regularly (Cascadia!).

Startup Riot is actually run by a team from Atlanta, with Sanjay Parekh playing the front man. The event is run as a pitch contest, including feedback from experienced entrepreneurs as judges, plus several keynote talks and lots of opportunity for connecting with other attendees. Here’s the description and agenda for the day.

I loved the event. It was well run, it was curated (no service providers allowed¹), it had great sponsors (MailChimp, Gist, Twilio, etc.), and there was a great crowd of people there. I was surprised to meet one entrepreneur who flew in all the way from the Ukraine to attend, but having attended one Startup Riot, I can definitely see why.

Vancouver Invades

There ended up being 10 companies from Vancouver attending. It was amazingly energizing to be in Seattle with this big Vancouver crew, and absolutely underlines that we should be spending more time working together to connect the Cascadia startup communities of Portland, Seattle, and Vancouver.

This was also the one area where I wish things had been run differently.

The Startup Riot team selects the startups that are allowed to present, and then swears them to secrecy: no one is publicly allowed to announce they are presenting. I heard about the event privately from a few startups from Vancouver who had been selected, but I never expected to show up and be immersed in a sea of Canadians.

I love & respect the cone of silence approach. It means that to find out more about the startups who are presenting, you have to make the effort to attend in person. And then the startups each have a booth as well as their pitch time, meaning you can spend time asking more questions and getting to know the team.

But the feedback I sent suggested that somehow the startups in each region be notified of who else was going.

If more people had known, I think the Vancouver area attendance would have been even higher, and would have let us organize group rides and shared accomodations and such.

The Presenting Companies

I tweeted about each of the companies whom I saw present, which ended up being the first 10 of the day, and the last 10 of the day. I missed the middle 10 pitches because I was in deep discussion with Red Russak, who labels himself a startup concierge and runs StartupSeattle.

I have tweets about the pitches I did see, as well as the two keynotes, over on Storify, as well as embedded as a slideshow here:

Pitch Quality

One thing I can say is that the pitches were for the most part not very good. Startup pitches are relatively formulaic, especially with such a short amount of time. I first heard the tip from Paul Kedrosky to cover three broad areas – why you? why this? why now? Show the product that you’ve built. Tell your customer stories. Say that you have paying customers (cough QuoteRobot cough).

The startup pitches I saw were missing a lot of these basic items. My sense is they were under-prepared, and had clearly not had much help / feedback from people who have done early stage startup pitches before. Of course I have opinions on which of the startups seemed to have “big ideas” or were likely to succeed. Others were more vicious in their complaints².

Ultimately, the story is about people trying to do their best to build a company, to get paid to be involved in solving a problem they are passionate about. Should all companies be big ideas? Should we all be building businesses focused on clean tech or nanotechnology? I’d rather see individuals doing startups of any kind – and learning from them – rather than working for a big company that is in need of replacing or at least reinvention.

No blame should attach to StartupRiot for this. All of those startups need to get better at explaining their story, at selling to customers, and about making people care. And they should ask for help, both from their community at large and from their peers.

StartupRiot was a chance for startups to stand up, tell their story, get some feedback from judges, and (hopefully) get some publicity and connect with people they wouldn’t have met otherwise. It definitely delivered.

See you in Atlanta

Congratulations to Sanjay and the entire team for putting on a great event. The next event is in February 2013 in Atlanta. I’m going to try and organize a group trip to attend. If you’re a Canadian startup and are interested, drop a line in the comments.

Footnotes

¹ This is what StartupRiot says about curating their attendees: We prescreen all applicants before allowing them into our events. For our main events that means we screen out all service providers. We do this because it leads to a better event for you, our sponsors, and us. I love curated events, and I think this is a great policy.

² There was a particularly harsh article written up in Seattle’s alternative news weekly, The Stranger. I captured my comments on my link blog.

Calendaring & Meeting Tools

2012-08-10T10:54:00-07:00

As I’m sitting here manually adding blocks of “break” time to my calendar around meetings I’ve booked, I’m getting a little frustrated.

I’ve used Tungle for several years now, and it’s quite a good tool. Rather than playing the are-you-free-on-Wednesday dance (or as I like to call it, “Calendar Tetris”) with multiple people, you can quickly find time when everyone is free to meet.

I’ve been worried about Tungle since they were bought by RIM. A bit of a face lift, but nothing much in the way of new features, and certainly no communications or marketing (the blog is frozen in time, to the date of the acquisition in May 2011). I think this a great ecosystem tool for BlackBerry - it addresses a core function of their user base: business users going to meetings.

Since I find what Tungle does very useful, I’ve started to look around for other calendaring / meeting tools that can solve Calendar Tetris.

I tried Boomerang Calendar for a bit, but it just copy-pastes free slots into email. That still puts the effort on me to juggle different slots in my calendar, and frankly the cut/paste is not that much of a win.

Adam Kalsey pointed out TimeBridge to me (acquired by MerchantCircle in 2010, no news on blog since then - uh oh). I’ve worked with it a little bit, but the fact that their meetwith.me domain is showing as expired AND hosted on GoDaddy doesn’t inspire confidence.

What I want is a calendaring agent. An agent that I can program to automatically add 30 minutes of blocked off time on either side of a meeting. A software agent that can know I prioritize meetings with significant others, paying clients, and long-time mentors. I’m seeing tools like IFTTT or Cloudsnap arise that let us program some of the many tools at our disposal. But I don’t want to program. I want an agent that learns from me, and improves automatically.

As my friend Alen likes to say, think about Surface vs. Purpose. The Surface: organizing meetings. The Purpose: staying organized and efficient, and living a life where you get more of the right things done.

I would love to hear suggestions for other tools that solve the Purpose aspect of calendaring. For now, you can find my availability online at tungle.me/boris.

Best practices for tracking QR Codes

2012-07-27T16:50:00-07:00

My friends at Common Craft have a book called The Art of Explanation coming out this fall, and it will include QR codes. In the book, there are multiple references to Common Craft videos, and Lee wanted to make it easy for people to go from reading a page in the book and then easily viewing the referenced video. These QR codes link to the explanation videos on their website. Since there is only one chance to get the codes right before they get printed, we talked about different ways to generate & track QR codes¹.

Short URLs and QR Codes

I’ve used a handful of link shortening and QR code generating services in the past, so we looked at which of them might be appropriate for generating codes and tracking. Link shortening services offer statistics on how many people have clicked on a link, which is a way to get tracking information even if you don’t own the domain you are linking to (and thus don’t have access to the analytics).

Many link shortening services also generate QR codes. Since the QR codes are generated based on the short URL, this also means that you can edit the link that the QR code points to after the fact. Useful if you have some QR codes in the wild in a permanent form (posters, t-shirts, books, etc.), but you want to link to different destination URLs over time.

There are two big downsides to using short URLs. One is relying on a third-party service, and the second is to add one more step between the QR code and the link you want people to actually end up on. If you are going to rely on a third party service, then you should at least use your own domain name. Then, even if the service goes down, you can run your own redirects, rather than risk having your QR codes break².

Lastly, while not really a downside, you’re not really tracking the the QR code, but rather the short link. This means if you use the link anywhere else publicly, you won’t necessarily be able to tell the difference between someone clicking on the link vs. scanning the QR code.

Here are three link shortening / sharing services that also create QR codes from those links:

bit.ly

While bit.ly has recently changed quite a bit, it is the original URL shortener that many people still use today. Using your own domain used to be a pro service, but is now free. You can go to the advanced tab of your bitly settings and set your custom domain.

Getting QR codes out of bitly is done by appending .qrcode to the end of any bitly URL. More details in the bitly documentation. Note: it looks like you can’t redirect your root domain unless you’re using Bitly Enterprise³.

Short Switch

ShortSwitch is a simple custom URL shortening service where the only option is to use your own domain. It starts at free for 1 user and 15 links per month, with the first paid plan being $4 / month for a single user account.

Each link you shorten also generates a QR code, and you can edit the URL the short link points to after the fact.

awe.sm

While awe.sm does shorten links, it’s design and strength is in social share tracking. It does generate QR codes, but these are done through the Google Charts API.

The basic account is $19 / month for a single user, and is definitely something you want to look at to get data on your social media conversion. Looking at it in more detail, it’s not the right solution if you are looking for QR-code related tracking, although it can be used to do it.

QR Code Tracking

The previous tools and the technique of creating short URLs with QR codes generated from those is more about tracking the links than tracking the QR codes. There are different ways to “track” the usage of QR codes directly.

Esponce

Esponce was a tool that Lee had found that is specifically designed to track various uses of QR codes. Under the covers, it actually does create it’s own short links for tracking links, but Esponce also supports generating and tracking different kinds of content, such as contacts, phone numbers, geo locations, etc.

Looking closer, Esponce uses its own domain for the short links it creates. This means that you need to either trust that Esponce will be around for a while, or to pay for a white label solution at $600+ per year so you can use your own domain for link shortening.

Still, there are lots of interesting features in this platform, and the white label solution could be very interesting for agencies that want to offer these sorts of services in house.

Tracking using Google Analytics

The solution we finally settled on was to track QR code usage by using Google Analytics.

Using the Google Analytics URL Builder, you can build. For Lee & Common Craft, the QR codes are going to be in a printed book, so we fill the builder out as follows:

Campaign Source: book
Campaign Medium: qrcode
Campaign Name: NameOfVideoBeingLinked

The generated URL will then look something like this:

http://www.commoncraft.com/video/wikis?utm_source=book&utm_medium=qrcode&utm_campaign=wiki

Once you’ve built your URL, you can use any QR code generator to create your QR code. Be careful not to use one that generates the code based on a short link, since the whole point is to remove that step completely. I recommend the ZXing generator, which is also great for things other than URLs⁴.

Now, when you use these QR codes out in the field, you know that people visiting your website using that URL are definitely coming from a scan of that QR code. If you continue to follow this pattern for all of your QR codes, you can run different campaigns and then build reports in Google Analytics to slice and dice in different ways.

The advantage of this approach is that you are not relying on a third party service of any kind. You need to generate QR codes once (which are always the same for the same URL) and they will work forever. As well, since you’re likely using Google Analytics to do reporting on your website traffic in any case, you can do all of your reporting including QR tracking in one place.

Why use QR codes?

I’m going to skip the long discussion of why-or-why-not to use QR codes. I think they have their place, and if you do use them, you should have tracking in place. I think the same goes for social media, and link shorteners and some of the tools I’ve mentioned above can help you figure out which of your channels are working for you.

Many people have suggested that educating people about QR codes is a waste, and that we should rather focus on just displaying web addresses directly. I’m personally fascinated by QR codes because they are one of the few tools that are available to us to link the “real world” with online. They work today, and let us experiment while Google Goggles, NFC, and other more elegant solutions become widely adopted by the mass market.

Hat tip to Monique at Boxcar Marketing who put us on the right track, and in general is a Google Analytics Jedi Master.

Footnotes

¹ If you want an easy way to explain QR codes, check out Common Craft’s video explanation.

² Dave Winer wrote about using meta-refresh to run your own future proof “static” URL shortener. Phil Windley has a more detailed write up using a Perl script and the correct Apache settings.

³ Bitly Enterprise is a completely different level of tool, with pricing starting at $995 / month.

⁴ The ZXing project (pronounced “Zebra crossing”) is an open source project for encoding and decoding barcode images of all kinds. You can use the generator online to create QR codes that support all kinds of content, including “just” URLs. There is also code available for you to use in any custom application where you need to generate or decode barcodes.

Static Site Generators Lightning Talk at HTML5 Vancouver Meetup

2012-07-08T23:26:00-07:00

I gave a quick 10 minute lightning talk at the HTML5 Vancouver Meetup group about static site generators (SSGs).

I ended up putting the presentation together using Hekyll, which is, itself, an SSG for making presentations using impress.js. impress.js is an HTML5-based clone of Prezi, the panning / zooming presentation app; I just opted for simple presentation mode.

Check out the SSG Lightning Talk or view it in the iframe below (use arrow keys to advance).

This presentation needs work (never mind the fact that using my new machine to do a presentation caused a bit of a fumble). I spent a lot of time futzing with the tech, which should’ve gone into tuning the content. I’m finding a bit of a problem determining the right level to cover with SSGs. There are a lot of moving pieces around generators, layouts, and hosting, plus trying to explore the “why” of SSGs and what kind of solutions they apply to.

Hekyll was an interesting experiment. I like that I can keep my presentations on GitHub directly. Theming needs work - or my design + CSS skills need work, depending on how you think about it :)

I also ended up demo’ing Prose.io (read Development Seed’s intro blog post), which is a web-based content editor for GitHub. So, if you’re creating / hosting content using GitHub’s native pages functionality (which uses the Jekyll static site generator), it means you can edit your content online in a friendly, Markdown-aware environment in your browser.

Northern Voice 2012 - Web Literacy

2012-07-02T23:01:00-07:00

I attended day 1 of Northern Voice 2012, including Moose Camp, which is a block of time that is run unconference style.

I hosted a discussion on the concept of web literacy. I went into the discussion thinking along the lines of being a web maker - being able to understand HTML and code. But as we tossed concepts back and forth, it’s clear that web literacy is a very broad term, and that there are different levels of literacy.

I’ve added a link to this blog post on the Northern Voice wiki – please add your own notes and links relevant to web literacy.

What does web literacy even mean?

I came into the discussion thinking about web literacy from a very how-does-the-Internet-work perspective. I envisioned people caring about the difference between a domain registrar, a DNS host, and web hosting¹.

The group quickly disabused me of that notion. Owning your own domain name and what the consequences of that are is several levels away from the basics.

The prevailing opinion was that web literacy begins with something more akin to media literacy: understanding where the information that you’re looking at comes from, knowing how to search, being able to differentiate between sponsored links and search results, and so on. Critical thinking about the source & context of the web page you are looking at, whether or not to trust that information, and perhaps some understanding of the bias of different kinds of web content – is this a news source, is this trying to sell me something, do I have some indication of what person or organization this page is being hosted by.

I think that’s an excellent starting point for anyone using the Internet.

Creating Content

But to me, even if we include these concepts as part of web literacy, it’s pure consumption. I think of the web as being what the name implies – a back and forth weave of pages and links, and unless you’re part of that weave, you aren’t web literate.

Blaine shared a story of how a friend of his was approached to help put a web page up. His friend turned to the person who asked this and said “You’re illiterate”.

Having moved off of my techie, protocol centric view, I feel good about saying that one definition of Level 1 of web literacy is understanding how to put your own content online. But this doesn’t mean hand coding an HTML page and FTP’ing it to a server somewhere.

I don’t think posting photos to Facebook or using an app to post to Twitter qualifies as putting your own content online. Creating a WordPress.com, Tumblr, or Blogger account and making your first post certainly does.

(Internet) Avengers, Assemble!

Brian Lamb described feeling himself getting shrill whenever he tried to explain why people should care about owning their own domain and hosting their own content. It simply doesn’t matter to the vast majority that their TwitPic photos have ads being sold next to them or that they don’t have their own domain name mapped to Tumblr.

I jotted down a few of the phrases that people said as we talked.

“A lot of people who could care, do care” – the fact that we had a group of people having this discussion at Northern Voice was heartening. To me, this says, hang out with your tribe who cares, but also look for opportunities to recruit people who want to care.

“There is a stigma against people who can’t read. Digital il-literacy doesn’t have a stigma” – almost the reverse is true, with people waving off the fact that they can’t do that “web stuff” or “don’t understand the Internet”. We should celebrate knowing more about the Internet and understanding web stuff.

While I have increasingly felt farther away from Northern Voice’s focus on the personal use of social media and blogging, this discussion really made me savour this end user-centric view.

Quantifying Levels of Web Literacy

Putting aside whether people should care about their level of web literacy – and about wanting to achieve a higher level of such literacy – I think some discussion about the different levels is useful.

Level 1: Web Media Literacy

This could just be media literacy with a focus on web media, plus search & browser best practices. This also means knowing what a browser and a search engine are, so I could be convinced to set my bar lower and make this Level 1 rather than the Level 0 that I want to assign it.

Level 2: Web Creator

An understanding of web pages and domains (typing in web addresses instead of just searching for everything). Knowledge of various blogging tools and platforms as a user, and being able to create new posts.

Level 3: Web & Internet Foundations

A basic understanding of HTML & CSS. Can edit HTML in the “view source” mode of rich text editors. IP addresses and domain names. Understands how to register a domain and then link that domain to a hosting account or hosted service.

Level 4: Self Hosting

Knows about web hosting, databases, and FTP. Can install and do basic setup of WordPress and other tools using a web hosting admin panel. Has access to a basic HTML editor and can create individual web pages from scratch.

Are these the right levels? This whole area seems like such a patchwork of systems, where you have to understand the whole stack and how it interoperates.

I’ve focused on web & Internet concepts, but completely skipped anything to do with images, video, or audio. Also skipped are anything related to web etiquette / legal topics, such as links as attribution, basics of quoting, fair use, copyright, etc.

Saving the Internet

I’m quite passionate about the Internet, mostly because of all the great people I’ve met that live there. Open source, open data, and just the basic human nature of communication, sharing, and connecting online are great things, and we should try and get better at it. Ironically, I’ve spent a lot of time taking online connections and trying to move them to offline in-person events.

I’m going to try and “level up” a few people, perhaps by running a Mozilla “Kitchen Table” event in the next couple of months.

I’d also like to continue the discussion about what web literacy means. How and where should we be teaching it? Libraries and universities both some like institutions that should be taking part in this. Are these types of levels useful? I’ve enjoyed Peter Rawsthorne’s explorations into Mozilla’s Open Badges – could we define and create badges for web literacy?

Echoing both Reilly’s Friday morning talk and Blaine’s Saturday keynote, the Internet does need our help. At the center, it needs civility, discourse, and education. At its edges, it needs wildness and experimentation.

Footnotes

¹Registrars, DNS, and Web Hosting: I keep meaning to write up a simple explanation of this, but it seems like a mini tutorial on its own when you consider diving down into things like IP addresses and sideways to concepts like domainers, typo squatters, and more. I just tell people to not use GoDaddy and use NameCheap instead and consider it a win for now.

Setting up a new MacBook Air plus Lion

2012-07-01T15:17:00-07:00

I’ve long used Norse Mythology to name my computers. The tiny but powerful 11” Macbook Air got named after Skidbladnir, Freyr’s collapsible ship.

Why I got the 11” Air

I was looking for a home computer. One that would sit on my desk and be used for personal projects and some gaming. I wanted something low cost but that would handle anything I threw at it.

I was considering a Mac Mini or an iMac as my non-laptop choices. Having a laptop means being able to use it for personal travel as well, although that was not a huge factor in making the decision.

The Mac Mini came very close, but the graphics card seemed a little weak. When thinking about using FaceTime or Skype, I realized that I’d also have to invest in a separate webcam, which was one too many strikes against the Mini.

The iMacs are great machines to have sitting on your desk. You need to get higher end versions to get the “better” graphics card, but they are all great choices. The more I looked at the different iMac configurations, the knowledge that there was an upgrade somewhere not too far away for the entire iMac line up meant it wasn’t the right choice.

So it was back to looking at laptops. I knew I didn’t need a fully tricked out machine, and that I had a 22” monitor sitting at my desk, so screen size wasn’t an issue. I also have a large 1TB external drive, so lots of storage wasn’t important either.

With all that, the MacBook Air 11” ended up being a great choice. I bumped up the memory to 8GB and decided to spring for the faster i7 processor, and I think it’s going to be a great machine. Being able to write this post with the machine on my lap on a crowded 257 bus headed to Horseshoe Bay just underlines what a nice portable machine this is.

Apps

I mainly do email, Internet, photo processing, and some web development tinkering with my personal machine.

I’ve had a BootCamp partition and run Windows on my last couple of my machines, pretty much exclusively to run Steam and Windows-only games. Since I’ve got Diablo 3 to suck up what little gaming time I have, I’ll keep this machine Windows free for the moment.

Most of the desktop apps I end up installing over time are mainly for experimentation – few of them seem to stick. I expect this trend to continue, although I think that “local” clients – either through sync or APIs – will appear more often as well.

The first couple of must have desktop apps that I’ve installed are:

Evernote - all my notes are here, available on all devices
Byword - since switching my blog over, I’ve experimented with different ways of writing Markdown; Byword has been great, especially in full screen mode, for long form writing
Dropbox - I’m lucky enough to have just over 7GB of free space. Aside from photos and code repos, all of my files are here.

My long time favourite basic programmer’s text editor is Smultron. It’s now gone closed source and available in a classic version and a Lion-only 4.x version. I’m thinking of attempting another try at learning & committing to a real UNIX-y editor. For now, emacs (the AquaMacs version) is what I’ll try.

Command Line

Ruby and Jekyll / Octopress are the first things I need to get up and running on the command line so that I can get this blog post published ;) As well as setting up a new machine, this is my first time on OS X Lion, so it was interesting to see what I had to do to get my development environment up and running.

Create / copy over .bash_profile to my home directory
If you’ve already installed XCode, then run Preferences > Downloads, click on Components and install the command line tools
Otherwise, just install OS X GCC Installer (which you may need for certain code builds regardless)
Install Homebrew, then edit .bash_profile to include export PATH="/usr/local/bin:$PATH"
Install newest Ruby brew install ruby, then edit .bash_profile to include export PATH="/usr/local/Cellar/ruby/1.9.3-p194/bin:$PATH"
Install Jekyll gem install jekyll
Install Octopress - follow the docs; I have the source already in DropBox, so just need to install the various dependencies via gem. You may have trouble installing rb-fsevent, in which case you’ll need this magic incantation: sudo xcode-select -switch /Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/
I’m hosting on Heroku, so I also needed gem install heroku

That’s it for the first couple of hours with the new machine. I’m sure there are some Lion-only apps and features that I need to dive into. Trackpad gestures + Mission Control + multiple desktops on this small screen is great so far.

*I’m very new to Ruby still – I ended up installing rbenv to manage different versions, but I don’t know that I’ll even need to get into that

The Posts that You Don't Write

2012-06-29T10:57:00-07:00

I had a great discussion yesterday where the phrase “the post I didn’t write” came up. This is a fantastic phrase, and led me down the route of examining why and when I don’t publish posts, and thinking about why others – especially in the context of working for an organization – don’t publish posts.

In one instance, I still haven’t finished writing my post discussing the Northern Voice web literacy session. The “I haven’t had the time” excuse comes up all the time, but the things we make time for and the things we don’t say something. I think my main worry is that I won’t do the topic justice.

I know I’ve been struggling a bit with this new blog. I have only a few posts here, they’re all quite long, and I post every 3 - 4 weeks. I have some other “posts I haven’t written” that haven’t made it here, for a number of reasons.

One reason is this pressure that the next post is “on message”. That it’s long, has links, and has some level of research, editing, and so on behind it. Should I be less emotional here? Should I write on my personal blog? or maybe my link blog if it’s not long enough (or not good enough)?

I am not a very personal writer. I have lots of technology opinions, thoughts about the evolution of tech & industries. I post a bit about cooking & recipes on my personal blog, but writing about feelings or doing introspection in public is not something that I feel comfortable doing, and do so very rarely.

When you’re attached to an organization, not posting has many other overtones. Is this story I’m telling mine to tell? Will other people in the organization agree or disagree with my version? Does telling this story expose too much? (again, that I’m not empowered or allowed to be telling).

Fear, uncertainty, and doubt lead to us closing up, to be less transparent, to protect ourselves by not exposing our thoughts. But not posting just means that you’re not opening yourself to help and feedback as well.

What posts have you not written, or are you struggling to write?

Reactions to Microsoft Surface

2012-06-19T11:20:00-07:00

I used Storify to collect a number of the reactions that have been ricocheting around the web in response to Microsoft’s announcement of the all new Surface tablet. I am puzzled by their pre-announcement, happy to see more competition in this space, and really hope they succeed in shipping.

Node based static site generators

2012-06-02T17:17:00-07:00

My first experience with node.js was following the ‘hello world’ tutorial on the front page, which I then extended to experiment with writing in Markdown and creating HTML pages on the fly. Not quite a static site generator, but a fun experiment in learning during the Mozilla Polyglot Hackathon.

I’m currently using Octopress to power this site as well as bmannconsulting.com (see my migration write up), but one of the things I’d like is the ability to not have to have access to my dev environment in order to publish pages. That is, right now I can create/edit Markdown files anywhere¹ since my blog source is in Dropbox, but to compile / publish it, I need access to a machine that has the development environment installed.

I am hoping to use a node.js-based static site generator running on Heroku or Nodejitsu to have the best of both worlds. A minimal http server to serve up the baked HTML static files, plus the ability to connect to a Dropbox folder with Markdown posts in it and bake them on demand.

My first stop was to look at the existing node.js-based static site generators. I was looking for something with the simplicity and elegance of Octopress. To me, that means simple, one file posts with included metadata plus simple, as close-to-HTML as possible templating. Here’s the list of projects I found, with a few notes on each one.

Wintersmith

Source Link: github.com/jnordberg/wintersmith
Last Updated: < 1 month ago
Pros:
- Small, discrete code base, including plugin architecture
- Under active development
Cons:
- Jade-based templates
- Written in CoffeeScript

DocPad

Source Link: github.com/bevry/docpad
< 1 month ago
Pros:
- Does a LOT.
- Under active development
Cons:
- Overkill - support for the many templating options plus using CoffeeScript makes it hard to start hacking on

Blacksmith

Source Link: github.com/flatiron/blacksmith
Last Updated: ~ 2 months ago
Pros:
- Has a generator and a server
- HTML-based templates
Cons:
- Metadata for posts in separate json file & each post in a separate folder
- Updated: incorrectly stated Jade-based templates - actually uses HTML-based templates

Scotch

Source Link: github.com/techwraith/scotch
Last Updated: ~ 3 months
Pros:
- Comes with a redis-based cache, but can also compile to static files
Cons:
- Needs geddy and redis running
- Couldn’t get it working

Wheat

Source Link: github.com/creationix/wheat
Last Updated: ~ 6 months
Pros:
- Serves files from a git repo
- Runs the the howtonode.org website
Cons:
- Couldn’t get it running
- HAML templates

Petrify

Source Link: github.com/caolan/petrify
Last Updated: 11 months
Pros
- Interesting JSON templates
Cons:
- Generator only
- Couldn’t get it running (outdated “require path” syntax)

Wintersmith comes closest to being what I want. It’s maintained, works well out of the box, and has a minimal codebase. But, the double whammy of being written in CoffeeScript and using Jade-based templates by default makes it a no go. There is a plugin for Swig templates, so perhaps I’ll keep experimenting with it.

After all of that, none of the existing node.js-based SSGs seem like a great fit for adding Dropbox support for. So, I’m going to attempt to write one myself. To recap what I’m looking for:

Site generator plus simple server for local previews + easy PaaS hosting
Minimal, close-to-HTML-based templating
Single file posts that include metadata
Easily hackable for node beginners (i.e. not written in CoffeeScript)
Killer feature will be Dropbox integration where a folder is watched and files are auto-published

I think the on-disk format and general feature set of Octopress is excellent, so a secondary goal will be to try and follow the guidelines of source files & metadata that Octopress supports as much as possible.

If anyone has pointers to code or libraries that might be a good starting point, please leave a comment.

¹ Markdown editors:

I currently have three different Markdown editors on my iPhone.

WriteUp (app): This is the app I started with. Great Markdown and Dropbox support. Cool new feature is support for Versions.
Byword (app): Focused on distraction free writing in Markdown, and that’s it. I’m also using it on my desktop for full screen writing.
Writing Kit (app): I’ve just added this app, which features the ability to do research with a built in web browser / search. I’ll likely be using this to do research & grab links, and then do the majority of my writing in Byword.

The New Hack Stack

2012-05-27T11:54:00-07:00

In the past, the starter stack for web programming was LAMP. The ‘P’ originally stood for Perl, and then became mainly PHP.

Today, with $5/month shared web hosting and thousands of PHP-based scripts & applications, this success is hard to argue with.

But the truth is, managing even a shared hosting account is hard, never mind an entire VPS. You need to know the OS, the web server, the language, and the database.

Revision control? Especially because of PHP’s ease of deployment and editing, revision control is an advanced topic. This leads to things like “just edit it on the server”, lack of updates, and even lack of upstream contributions.

So what do we teach beginners? what is the stack we should be promoting to web makers?

I think the new hack stack is already here. Git, Ruby or Node, and a PaaS for hosting.

Revision control as best practice

Git - and more specifically, the social community around GitHub - teaches both revision control, working together on code, and the concept of contributing back.

Rather than teaching revision control as an advanced topic, we should teach it as a best practice from day one. Part of the kata of code should be your own local commits. this also means you have a built in safety net from ‘wrecking’ or ‘screwing up’ your code that causes anxiety for beginners.

The availability of graphical clients for Git that are highly visual and actually quite friendly to use is another bonus. Even Windows users now have a great option in the GitHub for Windows client.

Ruby and Node are the new ‘P’

I am singling out Ruby and Node as beginner choices for several reasons.

One is clearly momentum. Ruby, and more specifically, the Rails framework, has gained huge mindshare and usage. Node has also been getting lots of mindshare, although the use of JavaScript and the evented model takes some getting used to.

Both Ruby and Node have strong package / library management. Finding and installing various features into your app is easy and fun.

Setting aside the install / setup of your initial environment, both do a great job of running locally. And this is built into the frameworks, and not a matter of configuring 2 or 3 other servers to get your code working.

Why not Python?

Python might rightly take its place as the third member in the middle of the stack. But my feeling is that it doesn’t have the same ‘beginner mind’ community as do Ruby and Node. This is likely because Python is an older language and has much broader application than being focused on web programming.

I think this is easily fixable with some blog posts, tutorials, and evangelism. Perhaps something the Django community could focus on.

The Power of PaaS

The different platforms often tout auto-scaling or various other performance-related features. but ease of deployment and no configuration required is what makes them perfect for beginners.

I remember having a Twitter discussion with Anil Dash, trying to explain that I thought that focusing on PaaS as the beginner option was a better choice than the pain of shared hosting. Specifically, my main point was:

…the mythical non-technical user who installs on shared hosting will usually get their fingers burned
@bmann twitter.com/bmann/status/…

I, personally, have now completely moved away from running any servers whatsoever. This could be seen as a downside – I can’t install arbitrary bits of scripts. But, I can deploy self-contained chunks of code / services as separate apps, which is arguably a better practice in any case.

I’ve personally had great experiences with Nodejitsu and Heroku. With something like AppFog supporting even PHP apps (they’ve got a JumpStart program), the benefits of PaaS can be applied to those apps which would have required shared hosting or a VPS in the past.

Hacking for beginners

The new hack stack makes it incredibly easy for beginners to get started to write their own custom apps. There are easier ways to get a CMS-powered website up and running, but for writing code with custom functionality from scratch, it’s hard to beat the killer combos of GitHub, Node or Rails, and a PaaS to host it publicly.

Services to Product

2012-05-07T23:20:00-07:00

Moving from services work to having a successful product is probably one of the hardest things to do. Well, perhaps no harder than doing any startup from scratch, but the benefit in doing it from a services company is that you have a built-in way to bootstrap yourself.

As a single consultant or small development shop, you work for a number of clients on different projects. You might become known as an “X” shop, where X is some particular framework or programming language, being the go-to team when the complex or large projects come up. You get to learn about different verticals and different problems, coming up with technical and design solutions to fit. And you make consistent income ¹ with a high flexibility around work hours and location.

With a product of your own, you get to invest in the things that matter to you - the design, the vision, the care given to the technical smoothing, or perhaps the user experience of contacting support. Really, you can invest in whatever parts of the product that you’d like. There is no client to wreck the design or insist on the wrong features or really, any of that other complicated dealing-with-clients stuff.

There is a fantastic recent post by the makers of Harvest, a time tracking app, that really speaks to the pride that a team can have in building a product:

Our team is comprised of people who believe in crafting beautiful code, sweating the details and pixels, and working hard to find the right words to express our ideas. At the end of each day, we want to take a step back and say, “that’s our best work yet.”
Harvest Blog On Craftsmanship

It’s somewhat ironic that a time tracking web app – likely an essential tool in the profitability of service companies – is referenced in the context of building a product instead.

Buying your time back

One of the things I’ve started using in my discussion with services companies is the concept of making the first $10,000 with their product. That’s $10K per month. I came up with that number as short hand for two intermediate developers working on it full time. That’s usually when it hits home that products are hard. Because I’m not adding in the cost of offices, or hosting, or, you know, stuff like marketing. The math for startup founders is a bit easier. Even in high-cost-of-living Vancouver, my standard back of envelope assigns $2K / month for founder salaries ². As a single developer, any product that makes you $2K per month means you have an unlimited runway. Do you try and build another product to see if it’s easier or harder to get customers? Do you add new features or add new distribution channels – you’ve got lots of options, since your base cost of living needs are met.

In any case, that first X dollar amount can be something that focuses your effort. In essence, you are buying your time back, by having customers who are funding your product.

Productizing Your Services

One of the discussions I have with many people is how to productize themselves and their services. This doesn’t apply just to software developers, but anybody who is being paid for their expertise on an hourly basis.

What I mean by productize is to have standard packages and pricing. Even if you translate those packages back into hours, this gives you the freedom to look at investing in tooling & automation – in building products inside your business that aren’t just hours.

Hourly billing is a fools game. It misaligns the motives of clients and consulting shops. You, as consulting shop, are driven to equate billing more hours with making more money. When instead you should focus on becoming more awesome at what you do, while continuing to rent that awesomeness to your clients. My friends at Denim & Steel call it “invention for hire”.

The switch to agile makes things a little bit better, especially with custom software development. You can set a flat rate per iteration for your team, perhaps with a flat rate “on boarding” with the client to set the features and timelines. An interesting new startup that is helping connect teams and clients who want to practice this method is Group Talent. I haven’t had any direct experience with the site yet, but anything that moves us away from RFP responses and hourly billing is a good thing.

Investing in your product

Every startup blog under the sun talks about customer development, lean, and various other ways to go through the process of building a startup product. This doesn’t change when you are doing it from inside a services company, but you should consider the time as an investment.

I started by talking about buying your time back. You should track the time you put into your product, just as if it was another client. If you’re lucky enough to have high margins on client work, you might just be able to buy back big chunks of time to invest. This could be a time period for the whole company, or having individuals dedicated to the product full time.

One note about building a product that scratches your own itch. If you are spinning out a tool that you are using yourself internally, you’ll need to work extra hard to be objective about the customers for the product. There are many many MANY project management, time tracking, and proprietary CMS platforms out there: why will yours succeed where so many have failed? As always, the test is will customers buy it? And I do mean “buy it” – having your clients use it as part of your consulting work is not a good enough test.

To get real feedback on whether this is a viable business, you have to actually attempt to sell it to people. It can be tough to flip the switch and attempt this. It also means that actual non-development hours (and potentially non-developer people!) are going to be needed to spread the word, get feedback, and generally dedicate time to making the product a business, with all that entails.

As a service company, you can trick yourself into carrying the cost for a product that simply isn’t making money. Setting sales goals – such as X paying customers after Y months – can be a good trigger to do further investment on your own, like dedicating a full time person before the product is bringing in a full salary. But, this can’t continue forever: the product has to be showing enough traction that your investment pays off.

No VC money required

I think this is one of the big things that people aspire to – to say they didn’t need a VC (or angel).

The recent story of MetaLab, a Victoria-based company, entitled Build the rocket first: From $0 to $500K in 1 year with no VC money had all sorts of attention. There are some great stats that make for interesting reference points:

three people were dedicated to the project for nine months
25% of the team’s time was allocated to client work
the estimate of the cost of their product, Flow, came in at $300K
there are now 10 people working on Flow full time

The article says “We broke all the rules”. Well, not really – I think they did all the right things in doing a services to product move. So the stories about the MetaLab team focusing is, I think, more about making a plan and sticking to it until they literally bought all of their time back (and more!).

Focus

I’ve been looking for an excuse to spread this post by Matt Henderson of Makalu Media more widely.

Matt wrote a post on How to schedule focus, suggesting that all of their projects get planned in blocks of a week. This means the entire team can be working on one thing - so everyone can be in this flow of having context and riffing off each others’ ideas, problems, and solutions. Versus the context switching overhead of juggling multiple clients, multiple project teams, and so on.

If you’re able to “buy” an entire iteration of work for your entire team, this would be the ideal way to move a product forward.

Consider this both a recommendation to follow more of what Matt & Makalu are up to, as well as a prompt for Matt to get us an update of how it’s working out!

Let’s Talk

I’ve been having all sorts of discussions with teams on both sides of the services vs. product equation. The title of this post specifically says “services to product” – since I believe that you should be investing some of your services time into trying to make a product.

I’ll be at the 2012 Polyglot Conference on May 26th, where I hope you’ll join me for a live discussion. Bring your own experiences and questions. Track the talk on the Polyglot Session Suggestions site, as well as adding your own comments and links.

Footnotes

1: I have heard some gripes recently that barely funded companies having their founders take home a “regular” salary of that $5K / month mark, or higher. I agree that that’s too high, barring special circumstances. Your payoff from a successful startup should be a lot more than a couple of thousand per month. With two founders, dialing back your take home pay means you can pay another person to be full time.

2: Most of @ASmartBear’s post focuses on the difficulties / painful realities, but he ends with this quote:

Most consulting companies don’t make much profit, and it’s one in a thousand that has the discipline to launch a successful product during off-hours. If you’re going to make it happen, you yourself need to be serious, disciplined, and relentless.
But you can do it.
And if you do, you’ve just self-financed a startup, made a nice living, mitigated much of the risk of product-only startups, and built a great team in the process.
Jason Cohen The Unfortunate Math Behind Consulting Companies

What I'm obsessed about

2012-04-25T22:42:00-07:00

Brad Feld is finding a lot of noise in the system, saying he is noticing:

…lots of drama that has nothing to do with innovation, creating great companies, or doing things that matter. I expect this noise will increase for a while as it always does whenever enthusiasm for startups and entrepreneurship increases. When that happens, I’ve learned that I need to go even deeper into the things I care about.
Brad Feld www.feld.com/wp/archives/2012/…

So, he identifies areas that he is obsessed about, and is going to dive deeper into them.

It’s pretty easy to tell what I’m obsessed about. It shows up in my tag clouds and the theme of my link blog. But it’s useful to reflect and broadcast these areas as well.

Here’s what I’m obsessed about:

death of binary documents: Evernote is a more powerful platform than Dropbox if you consider that native apps aren’t needed at all; Evernote is a giant distributed database with some layers on top of it that make it look like a notepad. Data flows seamlessly and can be mutated easily. Binary docs are flies in amber. This is a long arc that will take a while to complete. A supporting arc is the move to paperless for all things.
collaborative flow: Flowdock, Hojoki, Grove.io, Ginger, and HipChat are examples of this; this means both real-time “chat” rooms as well as connecting in various bots and agents to feed information and alerts into an always-on challenge. Tools like Yammer also fit, but are meant for company wide usage, while the tools I list are targeted more at dev teams
re-invention of email / inboxes: we can’t quite leave email behind, but look for ways to either offload traditional email comms (like the previous “flow” tools), or different approaches to “the inbox”
signal vs. noise: this is a long time obsession. It’s a common phrase, across industries, and across activities. These days social is adding to both sides of the equation. While I think algorithms are doing ever more interesting things in this space, I like the concept of curation, agents, and simply better tools for people to use directly (again, the previous two likely connect into this as well)
ebooks: the rise of the media rich EPUB3 format plus the social adoption of digital reading plus the recent move to DRM-free ebooks is opening up a lot of interesting opportunities
business data platforms: we see many small businesses or departmental teams adopting web SaaS tools; but, start looking at multiple tools and it becomes much more difficult - how do we not re-enter data? how do we create workflows that span multiple tools? how do we manage identity & billing across apps? we have several early solutions in this space, but there is only going to be an accelerating need for these types of tools. Chris Devore has written about this as well.

I probably can’t finish off this post without admitting that I’m still obsessed about startup ecosystems, specifically what it means to build more (and more successful) startups in Canada and Vancouver.

Vancouver tech needs to wake up

2012-04-25T14:35:00-07:00

Brent Holliday writes about what is needed for BC’s technology industry to succeed, which is itself a follow on to Jevon’s StartupNorth post on Canada’s next 5 years. I have some further thoughts on the five areas that Brent discusses:

Education

I have heard this “start ‘em young” story before. But I don’t think this is the battle that the startup tech community should be fighting. Absolutely we want web-literate youth, and Mozilla (as one example) is an organization that is doing amazing work here. But the pain the startup community faces is talent.

Let’s build more hackers. Intern, mentor, hire juniors. Code (and design) as craft, and move people up the ladder (rather than moving away because you’ve hit the ceiling in Vancouver). Hacker School is something that New York is doing that we could emulate locally. Promoting a startup as your first stop after university. Connecting with university co-op programs. But this is most appropriate for companies that are already heading up their growth curve, of which there are few at the moment.

Community as the framework

I agree with Brent a 100% here. But the web entrepreneurs don’t yet have a voice that is listened to by policy makers. Again, not something I feel like fighting, so let’s just double down and continue to work together without creating unnecessary bureaucracy. Let’s try lots of things and do more of what works. If it isn’t sustainable or doesn’t work, it’s just not meant to be. Danny Robinson’s proposed model for BCIC, with a UserVoice forum of the community voting for initiatives, and a model of early small investments leading to long term sustaining investments is perfect. Let’s clone that out in the open in the community, not run through some government entity.

Tighter Silicon Valley links

I think Vancouver is doing a good job here, but I don’t think Silicon Valley is the right answer. Rather, let’s build tighter links all over the world. Turkey, Romania, and the UK are three areas that have come up in recent conversations. We can be an excellent launch pad for those countries to connect with the North American market, as well as a convenient point to connect with Silicon Valley.

Policy

I see this as a supporting piece to make Vancouver-as-landing-pad succeed.

Grow like hell and don’t stop

The message is, make big bets and don’t cash in early. Still a tough thing to see for people going through their first startup loop. I believe we should be doing it, but it’s currently harder than it needs to be to follow this path. I don’t see enough Canadian investors on board with this model.

Brent ends with: “Shout about it or be proactively involved through the various organizations that are available. Right now”. Well, vantechia is my attempt at shouting about it. None of the current organizations are ones that I feel represent me as a web entrepreneur, so there is nothing that I can get behind and push. So, I’ll keep helping to grow companies, and spend a little bit of time shouting.

Octopress all the things

2012-04-24T15:20:00-07:00

I need to migrate my old blog posts off Posterous. I need to migrate my Drupal off Drupal. All will likely end up here.

Current thinking is that I will run three sites:

bmannconsulting.com - likely to be a simple landing page + a static exported-to-Octopress archive of the current BMC; perhaps hosted on Amazon S3 via rsync
blog.bmannconsulting.com - runs Octopress, hosted on Heroku (this site); long form posts, project pages, and so on.
links.bmannconsulting.com - running on Tumblr, high volume links + short commentary

Err… three sites for the main, public “tech” brand of me, which is the 10+ year old bmannconsulting.com domain. I have recently moved bmann.ca to Tumblr as well.

I’m going to likely outsource some of the work of moving posts around. I need a Posterous-to-Octopress script. The archive of Drupal posts need to be turned into markdown files as well. Not sure where to put those, might just go onto Github Pages as well.

Update: OK, I’ve got this running nicely. I’ve got a git checkout out sitting in my Dropbox, meaning I can use WriteUp or Byword to create / edit posts on the go. But, I have to be at a computer that has both my Dropbox and the full Ruby stack setup to generate the Octopress files.

So: how do I save updates to Dropbox and then have those updates generated and pushed to Heroku? I’m thinking it involves running another server just to do this and then involve the Dropbox API in some way. Thoughts?

Also, I have a job post on oDesk to write a Posterous Export - if you know anyone that could do this, point them at the listing, or look at project page.

Build more startups in Vancouver

2012-02-06T00:00:00-08:00

Last Thursday, Jesse Heaslip wrote ‘Two Problems with (Vancouver’s tech community) and Three Ways to Fix it’. If you don’t know Jesse, know this: he’s crazy-irrational-passionate about the Vancouver tech community. For proof, I offer the fact that he organized 40 tech events that were attended by 1600 people in 2011. Talk about impact!

Jesse lists the two problems as:

1. How do we get companies to the stage where there is interest from acquirers?

and

2. How do we get those companies to stay?

First of all, for a really great backgrounder on startups in Vancouver and the various funnels that are in place, read Greg Aasen and Danny Robinson’s post on the BCIC blog ¹. Money quote:

The only way to get more anchor companies is to start more companies. Some fail, some exit and some anchor. There is no shortcut.

For Problem #1, I first want to re-state the problem as “How do we get more companies to the Series A and/or acquire stage?”.

I added “more” to the statement, since I think that’s what we want.

Secondly, focusing on only acquirers is probably the wrong goal. Series A (before everything turned into super seed rounds etc.) used to be the first step of going big. Large VC funds² earmark millions not only for Series A, but to support companies in their Series B, and so on. We definitely do want anchor companies that grow big right here in Vancouver.

Jesse’s answer is “build a better community”, phrased as weaving the threads we have together very tightly. Allen Pike’s response focusing on homes for startups goes one step further and suggests shared spaces as the tactic to get us there.

I think the answer is simpler, and more like web traffic or a funnel as Greg and Danny wrote about: build more startups. Will a tighter community help us get there? I’m sure that is part of the puzzle.

Problem #2 is “How do we get them to stay?”. Jesse’s answers here are really three problem areas with suggestions on how to work on them. Academic-industry ties, #WeAreYVR community pride, and Hack Hut (a new coworking space).

I’m not convinced on school / industry issues, other than promoting startups as a place to work for new grads. As far as I am aware, new grads are not the main creator of new startups. As a counter-example, instead of accepting offers with Google / MSFT / Facebook etc. Cristian & Mircea turned down offers with those companies to do their own thing.

However, A Thinking Ape took those offers after graduating in Canada, moved away, then did their own thing, and then ended up in Vancouver. Time will tell if they stay.

I left after graduation in 1999 because the local opportunities that I could find in Vancouver were small and sleepy. I came back in 2004 because it was my home, and it just so happened that things were looking up.

Do we need to raise the profile of startups as a career option to keep some of this great talent in the local pool? You bet!

WeAreYVR. This seems to have the same goals as #MadeInVan. As I said in my blog post in response to that effort - great idea, how about we not re-invent the wheel - AGAIN.

This has been done badly again and again and again. A non profit entity like Vancouver Is Awesome could be the long running entity that might make a good home for this effort. But focus on the brand, on telling the story of local companies, not on the directory-done-badly.

On Hack Hut. Allen’s comments on this are excellent. Allen mentions The Hive. They look like an interesting organization, and I think working with them might accelerate this initiative. Also as Allen said, WorkSpace was the first hub, and it worked. I wish it was still here. Let’s vote with our dollars and support the spaces we already HAVE before deciding we need to create a new one.

Lastly (still on problem #2), I also don’t think “getting them to stay” is an issue. We need them to COME BACK. Stewart Butterfield and Avi Bryant are coming back. A Thinking Ape came back. We need experiences (and money and connections) from elsewhere to come back. So we can close the loop and do it all over again.

OK, now that I’ve made lots of commentary on Jesse’s post, let me say … I agree with him!

I think working together to raise awareness of the community and having shared spaces are the two strongest areas where the community itself can make an impact.

Other than that, the only thing I can suggest is to build more startups in Vancouver.

Footnotes

¹ It looks like the original Greg Aasen & Danny Robinson blog post has been truncated / removed / edited. I am going to try and find the original, since it clearly spelled out the stages and support systems of the BC tech ecosystem.

² I had a brief chat with someone from a large VC fund the other day. Accelerators and the collection of angels and super-angels around them are often focused on acquisition as an outcome: turning a couple of hundred thousand dollars into a couple of million dollars is a big win. For large VC funds, they NEED companies that want to go it alone, so they can put in millions that turn into 10s or 100s of millions.

I give larger VC funds a bad rap most of the time, but in this case the goals are aligned: they want companies to go big.

May 4, 2013: This was first posted in February of 2012. In the meantime, Hack Hut is now Launch Academy, which is over a year old. It is a non-profit, and so a great home for all sorts of initiatives. My Reinventing the Wheel One Directory at a Time post has also been updated and has more details.

I want to vote with my dollars that the Vancouver tech community is important

2012-02-05T09:01:00-08:00

Allen Pike posts about how important homes for startups are and what his experience has been like here in Vancouver. Here’s the closing sentence:

Bringing startups close to one another is, dollar for dollar, more helpful to the Vancouver ecosystem than tax breaks ever could be.
Allen Pike www.allenpike.com/2012/…

Once again, I realize that my gut agrees with this 1000%, and it has for years.

In May 2006 I wrote about what I called the Innovation Commons; at the time, I didn’t think coworking / community spaces could / should be for-profit. Last summer I referenced that post again in writing about coworking and supporting it with dollars and wrote this:

Vote to have something exist by supporting it beyond a retweet: with blood, sweat, and tears, or with cold, hard cash.

I think shared spaces are fantastic, and deserving of your dollar votes.

We are lucky to have spaces like The Network Hub and The Hive. If the economics of those spaces work for you, then pay for them.

If you can find a shared space and shack up with other startups: do it.

The article I was riffing on said “You cannot make a profit selling community”. And there is a whole bunch of other stuff in there about what coworking MEANS.

So, all of us that have an irrational desire to have the Vancouver tech community be, and happen, and grow - we need to go beyond the dollars.

I’m thinking back to an idea that we had in the early days of the Bootup Society, about a membership based program. About paying to have the Vancouver tech community exist.

I’m saying right here: let’s do this. I’m in for $50 / month or $500 per year. Maybe you can let me into some sort of space on evenings or weekends, but really, I just want to vote with my dollars that the Vancouver tech community is important.

Update May 1, 2012: I’ve ported this piece over to my new blog. At the time, Hack Hut was not yet in full swing, and I didn’t mention VHS.

Reinventing the wheel one directory at a time

2012-02-02T00:00:00-08:00

I just saw this GrowLab post about an initiative to get local companies to get listed and link to a directory of Vancouver companies.

Now, I definitely think we can do more to celebrate being based here in Vancouver, as we digital creatives go out to sell to the entire world, but I hate seeing us once again starting yet another directory from scratch!

What other directories could we re-use the data from?

How about Crunchbase (143 companies listed)?

Or Angel List (172 startups listed in just Vancouver)?

Or the StartupNorth Index?

Even Techvibes has a company directory!

Now, of the four, I think only CrunchBase has a Creative Commons license that easily encourages re-use of the data. (And yes, I’m still waiting for StartupIndex.ca to publish what their license is, and I’ve bitched … err, “suggested” … to Techvibes that they should have an open license, too).

But it means that companies have an incentive to fill out and keep the information up to date. Both StartupIndex and CrunchBase are wiki-style, meaning anyone can help keep the information up to date.

So, let’s re-think creating directories from scratch. Talk to some of these existing sites, and by all means make http://www.vanmade.ca an interesting hub to tell the stories of these companies.

But more stale directories is not what we need.

(ironically, this might have been just a brief comment on the GrowLab site, but comments are turned off)

Footnotes

May 4, 2013: I originally posted this in February 2012, so the listings of startups in those other directories has only gone up.

StartupNorth did the right thing and now have a CC-By Attribution NC License on their index (still need to have that listed somewhere other than a comment, and non-commercial is less free than just going ahead and making it by attribution).

My pick for a startup-focused open data repository is StartupGenome (Vancouver list here). Their mission on their about page perfectly syncs with my thoughts on city/region level startup data:

We want to enable local startup communities to collect, curate and display their city’s data anyway they want.

I’m working on a couple of things with Vancouver startup open data and Startup Genome, and had a super encouraging meeting with the Vancouver Economic Commission. Stay tuned!

Boris Mann's Blog - Future of web, mobile, and retail

Vancouver's talent are like raw resources

The Flickr Question

OpenPhoto

Self Hosting

What’s the answer to the question?

Footnotes

Working with Contractually

Connection Karma

Back to Contractually

Footnotes

StartupRiot Seattle 2012

Vancouver Invades

The Presenting Companies

Pitch Quality

See you in Atlanta

Footnotes

Calendaring & Meeting Tools

Best practices for tracking QR Codes

Short URLs and QR Codes

bit.ly

Short Switch

awe.sm

QR Code Tracking

Esponce

Tracking using Google Analytics

Why use QR codes?

Footnotes

Static Site Generators Lightning Talk at HTML5 Vancouver Meetup

Northern Voice 2012 - Web Literacy

What does web literacy even mean?

Creating Content

(Internet) Avengers, Assemble!

Quantifying Levels of Web Literacy

Level 1: Web Media Literacy

Level 2: Web Creator

Level 3: Web & Internet Foundations

Level 4: Self Hosting

Saving the Internet

Footnotes

Setting up a new MacBook Air plus Lion

Why I got the 11” Air

Apps

Command Line

The Posts that You Don't Write

Reactions to Microsoft Surface

Node based static site generators

Wintersmith

DocPad

Blacksmith

Scotch

Wheat

Petrify

1 Markdown editors:

The New Hack Stack

Revision control as best practice

Ruby and Node are the new ‘P’

Why not Python?

The Power of PaaS

Hacking for beginners

Services to Product

Buying your time back

Productizing Your Services

Investing in your product

No VC money required

Focus

Let’s Talk

What I'm obsessed about

Vancouver tech needs to wake up

Education

Community as the framework

Tighter Silicon Valley links

Policy

Grow like hell and don’t stop

Octopress all the things

Build more startups in Vancouver

Footnotes

I want to vote with my dollars that the Vancouver tech community is important

Reinventing the wheel one directory at a time

Footnotes

¹ Markdown editors: