Founder's Blog - Jitbit

Skype spying on you in Russia, China and who knows where else?

noreply@blogger.com (Alex) — Thu, 23 May 2013 19:03:00 +0000

Russian newspaper "Vedomosti" - a reputable Russian periodical co-founded by Dow Jones, The Financial Times and The Wall Street Journal - has reported that both the national security agency and the police are able to tap Skype conversations without even filing a court order.

Since I'm originally from Russia, I will translate the article for you:

03/14/2013

"Russian security services can tap Skype conversations. According to major IT-security players, the national security agencies including both FSB (Russian abbreviation for "Federal Security Service") and the police have been able to monitor Skype conversations and track user locations "for a couple of years now", says "IB-Group" CEO Ilia Sachkov - "This is exactly why our employees are not allowed to discuss any business matter on Skype".

"Since its acquisition of Skype in May 2011, Microsoft has added a legitimate monitoring technology to the tool", says newspaper's undisclosed source in Kremlin. "Now any user can be switched to a special mode where the encryption keys are generated on a server rather than on the user's phone or computer". Microsoft has been providing this technology to security services across the world, including Russia.

The two information security experts interviewed by the newspaper claim that access to Skype conversations is often granted to the secret service without even a court order. Sometimes it is simply provided "upon request". The newspaper's source in the Interior Ministry (simply put - the police) confirms that monitoring Skype users is not a complex task for a law enforcement agent in Russia.

The official representatives of the Interior Ministry and the Federal Security Service have refused to comment.

The newspaper has also interviewed two Russian entrepreneurs, who had to move to London (UK) because of the pressure from the authorities - and they have also confirmed that Skype can be intercepted: "Skype was invincible back in 2009, but it's not any more"

That's not the first time

That's not the first time we hear about Skype breaching privacy.

There's been reports that Skype is being listened to in China
There's been reports that Microsoft has been replacing Skype "supernodes" with Linux boxes moving from p2p-architecture to a more traditional one.
It's been discovered that some web-bot with a Redmond IP-address crawls all http-links you post into a Skype conversation (here's the source translated from German).
Jeffrey Nokel from the University of New Mexico has discovered that the Chinese installer even comes with a keylogger that "listens" to some specific "bad" words and combinations and sends them to the secret service later...
Skype leaks your location

And the list continues.

Just so you know.

Don’t Choose a SaaS Helpdesk! Really?

noreply@blogger.com (Alex) — Sun, 17 Mar 2013 08:47:00 +0000

This is a guest post by Dean Mitchell about the objections people raise against on-demand help desk apps. And since we're a saas help desk vendor we figured it fits our blog just perfectly.

It’s obvious to everyone but a crazy hermit living in a cave eating dried frogs and drinking unsavoury and unmentionable liquids that a helpdesk solution of some kind is an unquestionable necessity for the smooth running of your business. You may already have one that causes you to tear your hair out and scream creative and colourful profanities due to its convoluted and counter-intuitive nature. Bin it now! Life is way too short for that crap!

Choosing a shiny new helpdesk – one that’ll make life a walk in the park - comes down to two basic choices – hosted or downloaded. To the vast majority the idea of a hosted SaaS (Software as a Service) helpdesk located in the cloud seems like a no-brainer and we have to admit it is the option we use, but there are people who feel differently...

“I want my data right where I can see it, thank you!”

Yeah, that’s fair enough. I like to keep my phone in my back pocket at all times because it feels safe, but that’s false security because I’ve lost two down the toilet doing that! In much the same way how could keeping your data on your own server ever be as secure as on a hosted server that uses high security data centers utilizing redundant backups, biometric locks, secure data transfers and multi-level encryptions? The answer is it can’t but, hey, it’s your business and if you trust every spotty oik you employ implicitly it’s no problem!

“I do want to spend all my spare time fiddling with the code - honest!”

We also accept that there will be a small number of programmers reading this who salivate at the prospect of getting their grubby little mitts on the code. To most that’s just more hassle than it’s worth, but if fiddling and tweaking is your bag then forget about a SaaS helpdesk – a nuts and bolts download is for you.

“I want My Customer Services Department Where I Can See Them!”

In that case the big advantage of a hosted helpdesk is lost on you because your support team will be just as effective sipping coffee or at the bowling alley with a cloud based helpdesk that is accessible anywhere there is an internet connection. The thing is we think the ability to deal with queries on the move is so important we added that feature to our server based version to. Just accept it – it’s way more important that your team are always there for you than always chained to the desk!

“I Don’t Like Change! We’ve Always Used an On Site Helpdesk”

Evolution catches up with everyone eventually – just ask those dinosaurs! The face of customer interactions and servicing has changed forever with the advent of social media. You may dig your feet in the sand but “We’ve always done it this way” is not a good reason to stick with a server based helpdesk unless you have a massive team of support agents in one building and a support policy that is harder to understand than Joyce’s Finnegan’s Wake!

As you can see some of these arguments seem more valid that others but we suggest that even if you do stick with a downloadable helpdesk think long and hard about updating it to one that’s moving with the times and offering more than headaches and frustration.

Julian Assange vs Mark Zuckerberg

noreply@blogger.com (Alex) — Fri, 11 Jan 2013 09:56:00 +0000

Hit the Tail: Marketing Hacks from Jitbit

noreply@blogger.com (Alex) — Tue, 11 Dec 2012 10:12:00 +0000

Over the last couple of years we came up with a number of great marketing "hacks" here at Jitbit Software, that include:

1. "Marketing Mondays"

Every Monday the whole team works on nothing but marketing. Nobody writes any code, no one touches the servers. No hacking, no debugging, no database tuning... Just marketing. Brainstorming on new ideas, staring at Google Analytics numbers, reviewing conversion rates, running A/B tests, working on website design etc. I ~~stole~~ borrowed the idea from Mike Taber who was kind enough to share it when we met last year.

2. "Metrics Day"

Every 1st day of a month the whole team works on nothing but our SaaS metrics. Since our flagship product (hosted help desk) is a SaaS app we've set up an epic Google Docs spreadsheet that calculates all the key metrics. We enter some of the numbers manually (monthly visitors, signups, cancellations, new customers etc etc) and the spreadsheet calculates the rest - the conversion rates (visitor-to-trial, visitor-to-customer etc), the churn rate (percentage of cancelling users), the engagement rate (how many users have tried "feature X"), growth/drop factors for all the metrics and many many other things. We spend the whole day figuring what's worked and what's not, what should we try next, and where there might be a problem.

3. "Content Obligation"

Every week every team member has to write at least one article for the website, targeting one specific "long-tail" keyword. This hack worked well for many months and it's the one I wanted to dive into details about.

Catching the long tail

You all know what "long tail keyword" means. 80% of the search traffic comes from the long tail. To get that traffic, your website requires a content-creation strategy that needs to be both continuous and scalable. That's why we first came up with the following strategy: once a week every employee finds a longtail keyword and writes an article on it. Not just some spammy keyword-stuffed text, but a short human-readable article like this one: "Cloud help desk" - not exactly an extraordinary piece of writing, but a nice and readable article targeted at humans, not just search engines.

This works wonders when you're bootstrapping. But this strategy is not scalable, nor it is continuous. Not every team member has the required SEO skills to find the right keywords. Not every team member is a good writer, some of them spend hours trying to come up with an article - wouldn't it be more effective to simply let them write their code instead?

We decided to outsource both the writing part and keyword suggesting to third-party services. And while working with freelance-writers is a well known business (we quickly built a CMS for our freelancers where they can log in, pick a keyword from the list, read hints from us and submit an article) outsourcing the keyword-discovering part was not. Until I found Hittail.com.

Hittail.com

Hittail.com is a keyword-suggestion service that analyzes your traffic and discovers some easy-to-rank-for keywords.

To use HitTail, you first need to install their tracking code on your website. HitTail will take it from there and start tracking the keywords people search to find your site across all search engines. These search hits are displayed in real-time (30 seconds refresh-rate), and the most popular longtail terms are added to the suggestions section. The idea is not new, I blogged about Lazy SEO back in 2008.

Hittail doesn't have many competitors, it's simply the only service like this. So let's go straight to pros and cons.

The pros

Easy to use - the interface is well-designed, the realtime "search hits" page is fun to watch, the actual referring links is something I miss in Google Analytics. Nice.
It works - despite of all the cons below, I was able to find dozens of new keywords and write some new content for it. I got about 5% visitors increase in a week. Not bad.
Built-in article writing service is a nice touch. Just click the "write an article" button next to a suggested keyword, and 3 days and $19 bucks later you get a custom article on the topic. You might find this useful if you have no experience in managing a team of freelancers on your own. Hittail's service is more expensive than hiring a writer yourself, but trust me - if you don't have any experience dealing with freelancers, don't have a proven screening/interviewing process to filter good freelancers from the bad ones etc. - it is cheaper to use Hittail's service. Outsource that headache (oh, that is a headache).
Simple and neat. And I miss that these days, when every little twitter-scheduling app out there has a ton of features and a dozen of checkboxes in the UI

The cons

Poor performance on "multi-product" sites. If you have multiple products on one website (if you're Adobe, Microsoft, Apple or... well, Jitbit) you're in trouble. Say, you want to discover longtail keywords for just one product. In other words, only a part of your website will have the tracking code installed - the pages that start with "site.com/product/*". And that's where Hittail goes crazy. It keeps complaining it can't find the tracking code on the homepage (of course it can't), it doesn't show any suggestions, it throws errors when you fill the "URL" field (because it is actually a "domain" field), etc.

Obviously Hittail is targeted at small startups, where "1 company = 1 product" is a typical case. Don't they want to go after big companies - banks, car manufacturers, software giants? Even tiny software shops like us choose this business model sometimes.

UPDATE: Hittail's support was kind enough to contact me on this issue and said, that though the app does show the error message about not finding the tracking code, other systems work just fine. And I can confirm that - keyword suggestions are coming in.
Unintiutive signup process (not really a "con", just a small glitch). After the signup you're presented with a "Login" dialog that prompts you for a username and password, along with the "register" link. But you have no idea what the password is. There was no password in the confirmation email, no "password" field on the signup form... Clicking the "register" link takes you back to the sign-up form. I spent several minutes searching for a password until I realized that this "login" form - is actually a "pick yourself a username/password" dialog. Not very obvious. I bet that's where their funnel loses a lot of potential customers.
No "new" keywords. As far as I can see, the service suggests keywords you already have some traffic from. This is disappointing, I'd rather see some new non-obvious keyword opportunities. The algorithm should be something like this: first, get a ton of suggestions from online keyword tools (AdWords keyword tool, Ubersuggest etc) then check the SERP for this keywords, and if the top 3 websites in the SERP are weaker than yours (in terms of Google PR, Alexa rank, trust-rank, etc) - then go for this keyword. If not - move on to the next keyword. Who knows, maybe some day I'll create a service like this myself...

Overall, Hittail is definitely worth trying. It almost certainly will give you some new long tail keywords you can go after.

I've been using them for only a week now, so I will get back to this after a couple of months of usage to let you know how they performed over time.

Non-obvious time-management hacks

noreply@blogger.com (Alex) — Wed, 05 Dec 2012 15:12:00 +0000

Have fun!

This is my favorite time management principle. Even though it's hardly about saving time per se, but still... Don't be afraid to have fun! It's OK to have fun. Don't feel guilty for having it. Having fun is not losing time. So go ahead, grab your snowboard and go to the mountains for a week. Have a dinner with your wife. Take your kids to the waterpark. Try skydiving. Have a motorcycle trip across a couple of states (I did this two months ago on a rented bike, and it was amazing). Play a video game. Heck, write some code, if that's fun to you!

Have fun. It's fine. Don't put it off for the "better days" - there's no such thing as "better days" when all the work is done and the goals are achieved. By that time your body might not be capable of having fun anymore. You're getting old, remember? Human intelligence peaks at age 26 and keeps declining afterwards. We all get dumber over time. So go out and have fun right now before it's too late.

"The most dangerous way to lose time is not to spend it having fun, but to spend it doing fake work."

is my favorite quote by Paul Graham.

Now, the actual hacks:

The "unsubscribe" button

Unsubscribe from all the newsletters and SMS-ads. Instead of just deleting an annoying message - take a second to click the "unsubscribe" link at the bottom and save yourself hours in the future. Unsubscribe from all those "Follow-suggestions", "Quora-digests", "Here's what's happening on Twitter" etc. etc. These emails seem so useful at first... They're not.

Turn off the "passcode" on your phone

Turn off the "passcode" when unlocking your phone. These little things add up, you know... If you're concerned what happens when you lose your phone - there's a "remote wipe" feature for both Android and iPhone (via the "find my iPhone" app). Also, there's the "sign out all my sessions" button in Gmail and many other apps.

Use a hands-free

This one looks obvious but I was surprised how much more you can do with both hands available. No need to buy a fancy bluetooth headset, simple handphones that came with your phone will do.

Set iPhone in "Do Not Disturb" mode

OSX Mountain Lion "notification center" also has this feature, by the way.

Take a break when you're stuck

..and get back to that task later. Most of us have a really short attention span and our productivity degrades after a couple of hours of working on the same task. Take a break, grab a coffee, chat with someone, go for a walk. Then get back to it.

Set you phone email to "manual check"

Oh, this one saves a ton of time. Email can be a productivity killer. Set aside 10 minutes every hour for email (30 minutes every 3 hours is even better). Only few emails require immediate attention, and scheduling your email work can be a tremendous drain on productivity.

Don't waste your "on-desk" time

Workspace is for work. When you're behind your desk - get the work done. Check emails and Facebook on the go, use Instapaper or another "do-it-later" service for interesting but distracting websites.

Learn to say "no" to your friends

"Hey man, you know someone who can help me with XXX?" "Hey man, I'm thinking of launching a web-service, would love if someone technical glanced at my plan"

I get a ton of these... Don't get angry, your friends might not know, what it's like to be a founder. Instead of getting angry - find a polite way to say "no". Since most of the time you're on the road anyway, a short "I'm on my way to the airport" or "I'll be out of the country for the next week" gives them an idea about your way of life. They'll understand.

Have a short list of "things to do today" every morning

Share your hack in the comments!

Google Search is only 18% Search

noreply@blogger.com (Alex) — Mon, 03 Sep 2012 11:55:00 +0000

I was recently testing some of the keywords and positions for our hosted help-desk app and it suddenly occurred to me that 80% of the page were not actually the search results. Check this out:

My brain got used to filtering the ads out, so it never popped into my head before... We are used to this picture. I actually had to get up from my laptop, grab a coffee and then glance back at my monitor from across the room to notice this.

ADs vs Results: area size

Now, we're all technical people so let's do the math:

The screenshot above is 1280x960 pixels (a typical resolution for a 13" wide-screen laptop and some older 15"s). My Mac has a 1920x1200 resolution, but still I prefer not to browse in full screen (actually, the only apps I run full-screen are the development ones - Sublime Text, Visual Studio, XCode etc.).
The search results take up 535x425 pixels.
Which makes it 18.5% of the window

I do understand that this stuff is resolution-dependent, but still... Only 18.5% of the screen is devoted to something that people are actually looking for.

Let me show you what "18.5% of the screen" really looks like:

ADs vs Results: UI elements count

Now, enough with the area size. Let's count the links - the clickable text UI elements.

The page has about 45 different links in total. Only 5 of them are the actual search results (I do not count the "sitelinks" - the sub-links shown under some results and ADs). Which makes it about 11%. Only 11% of the total links on the page are the actual search results.

(If we do include the "sitelinks", it makes 57 links, 10 of which are the results, which is 17.6%).

OK, let's be fair, some of the links are the tools ("Google Docs", "Gmail"), some are search modifiers ("Search near...", "Search images"), some are Google-Account utilities ("Sign in", "Settings") so let's drop these links and buttons... Let's count only the "blue stuff". I.e. links that "look like" the search results, not including the "sitelinks". We have 18 links in total. 5 are the results. Which is 27%.

The bottom-line is: even dropping all the "secondary" UI-elements, the Ad/Results ratio is almost 4-to-1.

Was it always like this?

Unfortunately, I was unable to find a screenshot of Google's result page back from the late 90s, but found some stuff from the 2000s. I did an image search for "large" images, dated "before 2008", searching for "google results", "serp page" etc. Obviously, I was not able to find a screenshot for these particular keywords ("saas help desk"), so I tried to find a screen with as many ads as possible. Since "saas help desk" turns out to be a pretty competitive term. Here's what I found.

The AD/Results ratio is 8 links to 7. Which is 47% of the links are the actual results.

Now, the area. The results are 779x595 pixels. The total size is 1108x790 (which is even smaller than my original screenshot). Which makes it 53% of the screen is taken by the results, more than a half (actually, even more, since it's a smaller screenshot and there are no more ADs below the fold).

What does this mean?

Google has cut down the results area by three times - from 53% to 18%. The company is obviously interested in people clicking more ADs (in fact, I believe that's also the true reason behind "Penguin" and "Panda") since it's the company's primary source of income...

But all I know is that in the early 2000s Google has become the #1 search engine because of the three things:

Relevant results
The speed at which they were served
The simplicity of the UI

Looks like they're dropping #3 now. Is Google becoming a Yahoo?

Visual Studio & ReSharper performance tips

noreply@blogger.com (Alex) — Mon, 25 Jun 2012 11:42:00 +0000

Another post for ASP.NET/C# developers reading this blog. If you think these posts do not belong here, please leave a comment, and I'll consider moving my development articles to a separate blog.

Everyone knows Visual Studio 2010 can be slow. Even on my laptop with 8 gigs of RAM and Intel Core i7 it gets sluggish. Especially when editing Razor MVC views - every keystroke takes 1-3 seconds to register.

Below are a few tips on how to make it a bit faster:

Temporary disable unused extensions enabling them only when needed. Code highlighters, source control providers you rarely use (AnkhSvn, for instance, can cause slowdowns), script minimisers... All the rarely used stuff should be disabled. It's under "Tools - Extensions".
Disable "Track changes" in "Tools - Options - Text Editor". This way Visual Studio will not display a colored bar next to the changed lines in your source files.
Disable "Rich client visual experience" in the "Options - Environment". Here's how that settings page should look: screenshot
Disable IntelliTrace (Ultimate edition only) - in the "Tools - Options - IntelliTrace - General"
Disable "Track Active items" - "Tools - Options - Projects and Solutions - Track Active Item in Solution Explorer". This will turn off cursor-jumping in the solution explorer whenever you select different files in different projects.
Disable ReSharper IntelliSense in the ReSharper options, keep the default "Visual Studio" IntelliSense mode.
Disable ReSharper "Analyse errors in whole solution" - a huge performance consumer.
Disable editing enhancements in ReSharper - in ReSharper options turn off all the boxes under "Options - Environment - Editor"
Install all the hotfixes for your Visual Studio. You will find the list at http://connect.microsoft.com/VisualStudio/Downloads
Install SP1, and even if you already have it installed - reapplying it can help fixing the lagginess. Visual Studio sometimes "repairs" itself, reviving some old unfixed stuff.

IT Designation Definitions

noreply@blogger.com (Alex) — Sat, 09 Jun 2012 15:07:00 +0000

Continuing the "IT-humor" series of posts (previous: What If Drivers Were Hired Like Programmers) here goes:

Definitions of Designations

Project Manager is a Person who thinks nine women can deliver a baby in One month.

Developer is a Person who thinks it will take 18 months to deliver a Baby.

Onsite Coordinator is one who thinks single woman can deliver nine babies in one month.

Client is the one who doesn't know why he wants a baby.

Marketing Manager is a person who thinks he can deliver a baby even if no man and woman are available.

Resource Optimization Team thinks they don't need a man or woman; they'll produce a child with zero resources.

Documentation Team thinks they don't care whether the child is delivered, they'll just document 9 months.

Quality Auditor is the person who is never happy with a delivered baby.

Tester is a person who always tells that this is not the Right baby.

HR Manager is a person who thinks that...
a Donkey can deliver a Human Baby - if given 9 Months

Author unknown. Found this on a picture

Microconf vs. Business of Software

noreply@blogger.com (Alex) — Tue, 29 May 2012 20:10:00 +0000

Being a long time fan of the Business of Software conference, I finally decided to try something else and went to MicroConf - a conference for self-funded startups run by Mike Taber and Rob Walling.

It's crazy awesome and here's why: apart from the great speakers lineup, Mike and Rob did a great job narrowing down the right audience. These are product people. These are my people. You could grab a random person from the crowd and it would be a passionate founder or a co-founder, running a self-funded, profitable software company/startup. Like Balsamiq, Woothemes, Freckle, FlairBuilder, YouNeedABudget, SoftwareVerify, CrazyEgg or DotNetInvoice. Or Jitbit. We have the same problems, we share the same pains.

That's what I kinda miss at the Business of Software. "BOS" is an amazing event, it's like TED's for software companies. Their high price tag was, I guess, a way to filter out "wannapreneurs" and limit the audience to the "real deal". But along with the "real deal" came the big vendors. Companies that sell ERPs to Wallmart and do government contracts. Companies that ship real-time OS'es for combine harvesters and firewalls for Goldman Sachs. Sometimes not even software companies. These people are well-fed and not interested. Their ticket was paid by their employers. When Patrick McKenzie is talking his crazy awesome stuff about A/B-tests and conversions - they are like "meh?"

At Microconf, on the other hand, I saw people shouting "Eureka!", opening their laptops and hacking new features for their websites right there inspired by the speaker's idea.

Still, BOS has better logistics (i.e. "not in Vegas"), less "wannapreneurs" and classy speakers like Seth Godin, Jason Fried, Clay Christensen... Like I said - it's TED for the software industry.

Now I have two favorite conferences. So meet me in Boston this October.

PS. Ironically, the best outcome I got from BOS last year was... a dinner with Mike Taber, one of the Microconf founders. He was kind enough to share his "marketing monday" strategy we adopted here at Jitbit, and can't be more happy about that.

Once there was a search engine

noreply@blogger.com (Alex) — Mon, 02 Apr 2012 20:16:00 +0000

This is a sad story of an Internet giant. Started by two guys in a garage it eventually grew to control over 80% of the Internet search market and practically owned the whole Internet.

It was innovative and agile. The stock market loved it - at times the company stock doubled in price within just a month. It was one of the very few surviving companies after the dot-com bubble burst.

The company was adding more and more awesome free services: from newsgroups to image-hosting. It practically reinvented email by launching a great free web-based email app. It introduced a blogging platform and even tried entering the social networking space. It launched a free website-hosting service, introduced its own instant-messenger/voice application, social bookmarking service and many others. It published lots of open-source frameworks, tools and APIs for developers.

The apps it didn't create in-house - it acquired from the competitors, continuing to expand its range of services.

The company acquired several marketing agencies and started offering "paid inclusion" - adding ADs to the search results. Some people even claim that the term "PPC" was invented inside this very company.

But the company kept adding more confusing menus, bells and whistles to its homepage.

The search results were getting more and more cluttered by more and more ADs.

New players came to the field. And many users - first the early adopters then the rest of the crowd - chose the newcomers over the giant.

Eventually, the giant collapsed.

The giant's name was "Yahoo".

"History does not repeat itself, but it does rhyme" - Mark Twain.

Automating Amazon S3 backups on a Windows Server

noreply@blogger.com (Alex) — Sun, 18 Mar 2012 13:58:00 +0000

Our helpdesk app (the hosted version) is being used in about 170 companies. Users have uploaded almost 150k (150,000) files. We have logged about half a million tickets. From about 100k users.

Obviously, we need to backup all this data. And obviously, a consumer product like Dropbox or SugarSync is not an option. We needed something fast, reliable & scalable. We needed cloud storage. So we chose Amazon S3 and I decided to write this simple step-by-step guide for anyone interested in automating S3 backups on a Windows server:

1: Create an Amazon AWS account

If you don't already have an AWS account - create it here, it's free. Amazon's "free usage tier" on S3 gives you 5GB free storage from scratch, so after registering, sign in to your "AWS Management Console", select the "S3" tab and create one or more "buckets" on the left.

2: Get your access keys

You will need security credentials to access your online storage from the server, so click your account name - "Security Credentials" - "Access Keys" and copy your Key ID and Secret.

3: Download "S3Sync"

"S3Sync" is a great free command-line application from SprightlySoft. It is .NET-based and even comes with the source codes. At the time of writing this post their website was down, so I published the tool on Google Docs here: S3Sync.zip.

The tool syncs a given folder with your S3 bucket. And the best part - unlike similar scripts and utilities it performs a "smart" differential sync that detects additions, deletions and file-modifications.

4: Write a backup script

Create a batch file and paste this code into it:

S3Sync.exe -AWSAccessKeyId xxxxxxx -AWSSecretAccessKey xxxxxxx -SyncDirection upload -LocalFolderPath "C:\inetpub\wwwroot" -BucketName YOURBUCKETNAME

The code above is pretty self-explanatory. Just replace the "xxxxxx" with your access codes from #2, "YOURBUCKETNAME" with the name of your S3 bucket, and "C:\inetpub\wwwroot" - with the folder you want to backup. Then create a scheduled task that runs the batch file every 24 hours, and you're all set.

5: Pricing

Amazon gives you a 5 GB free storage during the first year. After the year ends the pricing is still very moderate. For instance, a 10GB storage will cost you about 60 cents a month which makes it a no-brainer.

Wait, don't add that feature

noreply@blogger.com (Alex) — Wed, 14 Mar 2012 17:28:00 +0000

Just had an argument with my partner. He was going to add a new logging feature to our help desk app. Another feature. Another button in the admin panel. Another setting. Another report to read.

This reminded me of Jamie Zawinski's law. The one that says every program attempts to expand until it can read mail.

Looks familiar?

Stop adding frigging features. If you think there's a killer-feature worth adding... think again.

If you're still positive about it - make sure that:

...the feature works exactly like it's supposed to.
...the feature works like the users expect it to which is even more important. If the user expects a fireman to resque his cat - give him a shiny "fireman" button, not a "cat rescue squad".

That's why Adobe Photoshop has so many photography terms in the user interface. The majority of Adobe users are photographers, who expect these names. They want menu items to say "aperture" and "exposure", instead of "blur the background" and "clip the whites".
...the feature is easy to locate. Add features, not easter-eggs. It won't add any value to your product if no user can find it. And adding a gigantic manual called "keyboard shortcuts and other hacks" is also not an option. The hacks, shortcuts and gestures should be obvious.
Don't make assumptions. Ask your users before adding a feature. Measure if the area of your app you're willing to improve is even being used by anyone. And that "genius" feature of yours might actually turn out to be a piece of crap. Lots and lots of startup founders I know keep avoiding this step.

PS. Image courtesy of Erik Burke from "stuffthatmatters.com", his site seem to be down, fortunately, had it saved on my HDD.

Why Developers Hate Antiviruses

noreply@blogger.com (Alex) — Wed, 25 Jan 2012 18:13:00 +0000

I hate antivirus software. I really do. Like almost every desktop software developer.

And the reasons are:

#1 - False-positive alarms

I'm sick and tired that my software is being detected as a "virus"...

...in spite of being signed with a trusted Verisign certificate.
...in spite of being .NET-based (a platform which is not very "virus-authoring-friendly", so to speak) and not even using any code-obfuscation.
...in spite of using the "ClickOnce" installation-technology (in other words - the code runs in a sandbox).
etc.

Why?

Because if your software has some kind of copy-protection built-in (encrypts and stores serial numbers, hides parts of the source code to protect from reverse engineering etc.) - an antivirus will most likely detect some "very dangerous" trojan.

Because if your software tracks mouse or monitors keyboard (like our AutoText for instance) - an antivirus will detect a malware.

Because if your software is some kind of a "compiler" - i.e. it's capable of building its own EXE-files - an antivirus will detect a self-replicating virus. Oh, and all your EXE-files will also be marked as viruses by the way (since you're most likely using a "self-executing-unpacker-code + data" architecture, which is considered a risk-factor by most antiviruses, no idea why).

Because if your software uses the "ClickOnce" technology (an auto-update framework that comes from Microsoft and is built-in to Windows!) - an antivirus will detect a "trojan downloader" and block your website in some cases.

These are just the few... And these are the actual reports I deal with every week. "Help, AVG blocks your installer saying it's a Trojan!", "Help, Opera has just blocked the downloaded file!".

EVERY. FUCKING. WEEK.

#2 - Antivirus vendors not dealing with false-positive reports properly

Now, dear antivirus companies! I understand - Users come first. Their security is your utmost concern. If someone sends you a virus sample - dealing with it is your #1 priority. I understand. I'm one of your paying users after all.

But please don't forget about us, the developers. We do send samples as well - the "false-positive" samples. We deserve some response. Fine, let it be within a week. Two weeks. A month. Two months. But please react!

Instead, some of you do not even have a feedback form or a forum on your website so we can upload a false-positive... And those who do, sometimes require us to send you our code-signing certificates, home-addresses, company papers and photo-IDs... We're guilty by suspicion. Everything is a virus until the author proves the opposite.

PS. In fairness, though, some antivirus companies do have these feedback forms in place, have nice developer support, and react promptly.

#3 - Antivirus my ass!

Create a simple C program with a code like this:

#include <windows.h>

#include <string>



int __stdcall WinMain(HINSTANCE,HINSTANCE,LPSTR,int)

{

  return MessageBoxA(0,(std::string("->")+GetCommandLineA()+"<-").c_str(),"Cmdline",0);

}

Compile it with a free "express" edition of Visual Studio 2008:
cl -Os -EHs-c- -GR- -MD test.cpp /link -fixed:no user32.lib -incremental:no -out:test.exe

Now test this program with your favorite antivirus. Voila! A "TR/ATRAPS.Gen" has been detected.

Congrats! You've just wrote your first virus!

[Infographics] The Ultimate Career Advice

noreply@blogger.com (Alex) — Fri, 16 Dec 2011 10:55:00 +0000

I'm SO happy to be in the middle.

And don't forget to check the 24 Steps To Success.

Rethinking the Cloud

noreply@blogger.com (Alex) — Mon, 28 Nov 2011 14:20:00 +0000

For months I've been thinking that a "cloud-server" is just an overpriced version of a VPS.

Don't get me wrong, I get the platform - as a software engineer. As a business owner - I even get all the benefits - elasticity, reliability, scalability, flexibility and all the other "bilities". On the other hand...

See, a regular VPS also runs "virtually" - just like the cloud. It's isolated from the host-machine failures - just like the cloud. The latest VPS-software (say, vSphere) can even do load-balancing, shadowing and real-time switching between physical hosts in case of a hardware failure - just like the cloud.

So, what's the difference then?

Actually with a VPS you might end up paying for unused CPU-cycles, unused hard-disk operations and unused extra RAM. But that's fine with me. As long as it's cheaper.

No cloud for "fat" servers

Let's have a look at one of our virtual servers - it runs a couple of static websites and two huge ASP.NET SaaS-applications along with their databases. It has an 80 gig hard-drive and 4 gigs of memory. Every month it eats up 500 gigs of traffic. And the monthly price is under $100. Now, according to Amazon's Pricing Calculator setting up a similar server in the cloud would cost around $500 a month. Seems expensive?

That's not how you estimate the cloud

Recently, while doing a project for our hosted help desk app, I realized that's not how you estimate the cloud. While VPS'es are fine for heavy & continuous work, the cloud can be great for occasional stuff, when buying a separate server would be ridiculous.

Real-life example: virtual emails for the helpdesk app

We figured we need a new feature for our helpdesk app. We wanted to provide our customers with free temporary email-addresses. And not just random addresses, but the ones that use the same subdomain as the SaaS web-application.

I mean, if "Acme Motorcycles" runs their SaaS-helpdesk at "acme.jitbit.com" I want to give them the "support@acme.jitbit.com" address. So they can start using it right away, without configuring POP3/IMAP interfaces, or any other complicated stuff.

To make this work I needed a "catch-all" email setup that will intercept all emails like *@*.jitbit.com. And the only way I found I could do this (aside from writing lots of code and setting up a weird DNS configuration) was to set up a Linux machine running Postfix agent.

I couldn't use my existing servers cause they're Windows-based. I did not want to rent a $20 linux box just for that... And then I realized that the cloud could be the ideal solution.

I'll save the step-by-step instructions for another post, let me just concentrate on the price here.

It turned out an Amazon server doing this kind of mail routing for us would cost... Around 35 cents a month.

35 cents a month. It even has a 10 GB drive (9.5 GB of which is obviously a free space) and runs 24/7. Which means, if I lower the disk space and set up a script that schedules this server to start/stop every 10 minutes (enough for email routing) it'd be even less. But I won't even bother. Cause - did I mention - Amazon's micro instances stay free for the first year.

I'm a cloud fanboy now.

Trying Rails & Mac, confessions of a .NET developer

noreply@blogger.com (Alex) — Sat, 12 Nov 2011 19:08:00 +0000

I run Windows on my MacBook.

Windows - because I'm a .NET developer and our startup is mainly Microsoft-based. And, to be honest, I like Windows 7. Finally, a decent OS from Microsoft.

And MacBook - because it's simply the best hardware you can get for it. The unibody design is awesome, the keyboard & touchpad are great, the 17" screen is fabulous... I love my Mac. I'm practically an Apple fanboy.

The only thing I don't like about my Mac - is Mac OS X.

It just didn't suit me from the very beginning. That blurry font-rendering, that mouse acceleration you can't disable... Also, I'm a gaming junkie and, let's say, Mac is not the most popular gaming platform.

But let's try a Mac for a change

Yesterday we had a new project idea and my cofounder suggested we do it on Rails. I was pretty open-minded about this. After all, it's always nice to try new languages, frameworks and environments, right?

And, hey, all the cool startup guys do their startups on Macs! There must be a reason, right? I want to be like that 37Signalish/YCombinatorish startup crowd. I want to be like those cool kids doing trendy Web 2.0 apps on their Macs. I don't want to be "a PC" from the Apple commercials. I want to be the "skinny-hipster-Mac". So let's develop in Ruby. Hell yeah!

Also, Macs should be easy to use, right? Everything works "out of the box", right? I mean, my Mac came from Apple. Same company that made my iPhone. And my iPad. And my iPod. And my other iPod (gee I really am an Apple fanboy). Also, I'm kinda familiar with *nix operating systems, the Bash command processor and stuff... So, there should be no problem right?

"Totally", - that was Max, my cofounder, - "no problem. Works right out of the freaking box"

The long journey

"Out of the box" turned into 40 hours of installing, reinstalling and configuring stuff. I'm not judging, just describing the steps I had to take to prepare my Mac to write some Rails code. Maybe it's because I'm just a dumb Windows guy.

First I figured that I need, well, Rails installed. Along with a Mercurial client. That should be enough to get me started and shouldn't take more than an hour to install.

While trying to install both Rails and Mercurial I got a bunch of error messages about an outdated Python version I had on my Mac. Fine, let's get a fresh version from Python.org... Didn't work. The "Pyhton.prg" version installs side-by-side with the built-in "Apple Python" and Mac OS still thinks I have an old version... After some head-scratching I decided to let my Mac check for updates - and bingo, Python was updated. But still, the version was not good enough for Rails & Mercurial... After another hour of black magic with python & homebrew and forcing another software update the Mercurial client was finally up & working.

But not Rails. It turned out it requires the "ruby-dev" package installed. And, well, the only way to get "ruby-dev" I found - is to install XCode. The big, heavyweight development suit from Apple.

Fine.

I tried downloading version 4.0 but the Apple website gave me an error in Chrome. I tried opening it in Safari, and it asked me to regster as an Apple developer.

Fine.

Let's get a developer ID. I filled out a lengthy form, confirmed my email address, got myself a developer ID, and... the Apple website showed me the exact same error message I got from Chrome earlier.

OK, there is another way to install XCode - to get it from the AppStore. But you need a Mac OS 10.7 Lion for that. Which is another hour of downloading, another hour installing and another $30 to Apple, because Lion is a paid sucka... OK, I'm on it. Gimme that Lion.

Another couple of hours downloading XCode 4.2, installing it, configuring all the "gems", and finally, after 40 hours of having sex with my Mac I was finally able to open "localhost:3000" and start writing some code.

Same on Windows

During that weekend I had to reboot to Windows a couple of times to fix some minor bugs in our helpdesk app and do some support stuff... And at some point, while in Windows, I had to make an urgent fix in our Rails app. I was too lazy to reboot back to Mac OS. So I decided to fix that on Windows.

I went to RailsInstaller.org, installed Rails, "pulled" the app from the repo, made the change, tested it on "localhost:3000", and pushed my changes back to the repo. And all that - in 9 minutes.

Wow.

9 minutes. Including the coding part. And including the time I spent Googling for tips & best practices on setting up Notepad++ for Rails development.

Sorry, folks. Call me a pervert, but I'm doing Rails development on Windows now.

I know!

I know that it's just me being dumb and, maybe, too used to Windows-ish ways of doing things. I know my Mac software was outdated. I know that every Mac developer probably already has XCode installed. I know I was having problems with the Unix-part of my Mac, not the shiny part of Mac OS. I know that in Windows it also takes time to install Visual Studio, SQL Server and other stuff to get you started (none of this is required for Rails though), I know comparing setup times on a platform you work with daily to a platform you've never used before is a bit unfair.

Some day I will make another attempt to move to the Light Side of the Force. Who knows, maybe even learn Vim! But we have a project to get done, so... Maybe next weekend.

Boosting Creativity Tip #1

noreply@blogger.com (Alex) — Mon, 31 Oct 2011 14:38:00 +0000

"Ideas don't come from watching television" Seth Godin.

Stop watching. Start reading books.

When you read a book, your brain is in the "uber-creative" mode. It analyses the text and tries to visualize what you read. It builds abstractions, pictures, sometimes even a whole new universe... Every time your read something - your brain is working out in gym.

On the contrary, when you watch - everything is pre-created for you. You just sit there and consume the picture. Your brain rests. Everything is pre-constructed.

And just like your body passes through different sleep-phases before it goes to the REM sleeping phase, your brain passes through different creativity phases before it reaches the "uber-creativity" peak. This means, the reading process should last, uninterrupted, 10 minutes least. So we're talking fiction books, not Twitter, not blogs or news feeds. Not even professional literature. Fiction books. Biographies. Semi-fiction books, like Steve Jobs story or Anything You Want by Derek Sivers (highly recommended, by the way).

All the coolest ideas I had - I had while reading a book or listening to a great talk. So here are some tips for reading more:

Take reading to where you have to wait - in the lines, at the landromat. Use public transportation - buses and the tube - instead of a car.
Use gadgets. Use your smartphone. Buy an electronic reader and take it wherever you go. Since I bought my first Kindle, I read much more. Much more. Though now I prefer Sony over Kindle.
Always have something to read. And contrary to common belief - it's great to read several books at a time. It's working out, remember?
Use audiobooks when jogging, biking or cooking
Read less. Read books you enjoy. If the book doesn't feel "right" after the first chapter - move on, don't force yourself. Don't feel pressured to read a book just because someone gave it to you as a present. Or you got it free at some conference. Stop. Move on. Otherwise only a half of your brain will be actually reading, the other half will be busy fixing your attention.

Outgoing Links Effect for SEO: Experiment

noreply@blogger.com (Alex) — Fri, 26 Aug 2011 08:37:00 +0000

There's been a lot of debate about whether external linking helps or hurts your SEO and most of the SEO experts including the gurus at SEOmoz tend to think of external-linking as a good strategy.

I made a simple experiment about six months ago. And discovered the flip-side of the coin. Here are the exact steps I took:

I have a page that ranks on top of Google's SERP. It's our Macro Recorder homepage. I've put a lot of effort to get it there and it took us five (5) years of hard "white-hat" work to rank high on "macro recorder", "mouse recorder", "macro program", "keyboard macro" and all the variations.
I've put an external link on that page - linking to a Wikipedia-article on the very same topic - "macro recording". It's a highly relevant link, obviously. And what can be more trusted and authoritative than Wikipedia?.
Boom. In one day the Wikipedia article has outranked my page in the SERP.

Enough said.

So be careful.

I guess in my particular case the reason was as follows: domains like Wikipedia are much, much more "trusted" and "relevant" in Google's eyes, that's why adding one tiny link to it's macro-recording article rocket-launched the article outranking my own page.

Removing the link restored rankings back to normal. But it took several days.

Fixing "Googlebot can't access your site"

noreply@blogger.com (Alex) — Fri, 19 Aug 2011 07:34:00 +0000

Got this email from Google about one of our websites recently:

Googlebot can't access your site

Over the last 24 hours, Googlebot encountered XX errors while attempting to connect to your site. Your site's overall connection failure rate is XX%. You can see more details about these errors in Webmaster Tools.

...

After you think you've fixed the problem, use "Fetch as Google" to verify that Googlebot can properly access your site.

The site was loading fine from all possible locations so I tried this "Fetch as Google" thing. Which resulted in an endless "pending" status that finally turned into "Failed":

I spent hours trying to investigate and finally found a solution. So if you ever bump into this, here goes

The cause

It was the old Google Analytics tracking code. Somehow it blocked Googlebot from indexing your website in some cases.

The solution

Just upgrade to the new, asynchronous tracking code and use the "Fetch as Google" tool once again.

Go Get A Cofounder [Mistakes I Made #5]

noreply@blogger.com (Alex) — Wed, 03 Aug 2011 14:46:00 +0000

This is the 5th post in the "Mistakes I made" series, where I share the "donts" of my startup experience.

I'll start from afar. My website has a number of pricing tables and I thought the tables look just fine. Until one morning I realized that they're a complete usability nightmare. Here's the "before" look:

A messy pile of text, prices, red-fonts and buttons organized into variable-height rows. I spent an hour or two tuning it until I had a much cleaner design:

A small number of tiny changes - making the price "stand out", adding background colors to the table rows etc - has improved my conversions by almost 10% and sure saved me from a number of support-questions... But I had a crappy pricing page for years. All because I didn't have a cofounder to tell me to "go and redesign this crap".

You need someone to call you an idiot

There are downsides of being a single founder. Even aside from the fact that a startup is too much for one person to bear. You need someone to challenge you. Someone to question your decisions.

Someone to call you an idiot. To tell you "man, this is crap, and needs rethinking".

Employees won't call you an idiot. You're the boss. Even if they try, they fall back as soon as they face resistance - "ah, whatever, you're the boss, it's your call". In fact, most of the failed startups I know have this in common: the lack of someone calling the founder an idiot. That eventually makes him come to believe that he's always right.

Customers won't call you an idiot as well. The ones that would - have already chosen your competitor's product. Also, as we all know from Joel:

Customers Don't Know What They Want.
Stop Expecting Customers to Know What They Want.

As a single person in charge you get used to your creations (that you're obviously in love with), not noticing the obvious idiocies. At that point getting a cofounder becomes more important than anything else. So go and get one now.

PS. I highly recommend reading PG's essay on this (scroll to #6).

"90% of your users are idiots"

noreply@blogger.com (Alex) — Mon, 27 Jun 2011 14:12:00 +0000

I just overheard this conversation between two developers at a co-working site:

"I plan creating a prototype for my new XXXX application, whatcha think it should be - a web-app, or a desktop app?"

The answer was:

"90% of your users are idiots who won't be able to tell the difference"

I think I just found my answer to the ultimate question of life universe and everything and it's not "42".

It's "90% of your users are idiots".

Now thats a questionable term to describe your customers. I don't think 90% of my users are idiots. But. That's the way to think of your users when making design decisions and building your interface.

Wondering if you should make this main button big or small?
"90% of your users are idiots"
Wondering if it should be one button or two?
"90% of your users are idiots"
Wondering if your installer should ask all those really important questions, like, how it should name the installation folder under "Program Files"? Or just perform the default action with no questions at all?
"90% of your users are idiots"
You're absolutely sure that the "pricing" page of your website should have fifteen different "plans" of your SaaS application?
"90% of your users are idiots"
Wondering if you should publish the installer as a ZIP-file or as an EXE-file?
"90% of your users are idiots"

And, in good conscience, don't you want to be an idiot when you're on the other side of the screen? I do! I want to be an idiot!

Please let me be an idiot. I want things to "just work". Don't make me figure my way through all the setup procedures. "Don't make me think" (c) Steve Krug.

I've been a computer nerd since I'm 11. I mean, I love the command line and stuff... But when I urgently need a port-scanner to test my server vulnerabilities after it's been hacked, I don't want the nmap tool, with a dozen of command-line options and a bunch of drivers it requires me to install. I just want a big red "scan" button. I'm an idiot, OK?

What If Drivers Were Hired Like Programmers?

noreply@blogger.com (Alex) — Sun, 22 May 2011 10:22:00 +0000

What if drivers were hired like software developers?

Job title: car driver

Job requirements: professional skills in driving normal- and heavy-freight cars, buses and trucks, trolley buses, trams, subways, tractors, shovel diggers, contemporary light and heavy tanks currently in use by NATO countries.

Skills in rally and extreme driving are obligatory!
Formula-1 driving experience is a plus.

Knowledge and experience in repairing of piston and rotor/Wankel engines, automatic and manual transmissions, ignition systems, board computer, ABS, ABD, GPS and car-audio systems by world-known manufacturers - obligatory!

Experience with car-painting and tinsmith tasks is a plus.

The applicants must have certificates by BMW, General Motors and Bosch, but not older than two years.

Compensation: $15-$20/hour, depends on the interview result.

Education requirements: Bachelor's Degree of Engineering.

Saw this on a programmer's forum, but was unable to locate the original. Let me know, I'd be happy to link.

Rootkit on a brand new Toshiba Laptop

noreply@blogger.com (Alex) — Sat, 30 Apr 2011 12:02:00 +0000

Sorry for the offtopic, this post has nothing to do with startups, web-development or entrepreneurship, but I felt I should still write this

I've just discovered a built-in rootkit in my wife's brand new Toshiba laptop. A non-removable malicious software application right from the manufacturer. That even captured and sent-out screenshots of my wife's work... But first things first.

First, let me apologize for the tone of this post and kinda incoherent writing. Please try to imagine where I am right now and please accept my apologies - I just finished dealing with this issue, like, 10 minutes ago. And, to be honest, I'm angry as a bear.

It all started with some corrupted files & folders on my wife's laptop. No problem - I launched the "CHKDSK" utility and scheduled a disk scan on restart. No big deal, right?

Except - there was no disk scan when I rebooted. I tried again - no scan. I tried everything: rebooting to safe-mode, marking the disk as a "dirty" one with the "CHKNTFS" tool, booting with recovery disk - nothing helped. I just couldn't launch checkdisk or schedule it for the next startup. So, I figured that the checkdisk file itself might be corrupted, so I ran "SFC /scannow" command that, supposedly, should restore it.

The command went up to 47% and aborted with the error message "Windows Resource Protection could not perform the requested operation". Hmmm... May be there's a virus preventing this? So I opened the Process Explorer tool (God bless SysInternals) and found a suspicious process called "rcpnetp.exe".

Why hello there! The process has no "Description" and "Company Name" fields, it loads "rcpnetp.dll" via AUTOCHK.EXE. A-ha! The tool that is supposed to launch startup disk scan! This can't be a coincidence. I opened Autoruns (God bless SysInternals #2) trying to find some registry key or something that launched this "rcpnetp" process.

Surprisingly, I found nothing. I decided to kill the process, delete those files from the "System32" folder and reboot the laptop. Imagine my frustration when those processes were back there, up and running!

I spent hours trying to figure, where this monster launches from... I tried several antiviruses, manual registry search, SysInternals tools... Nothing. So I turned to Google. And found some links (the second link is in Russian).

It turns out the files are loaded from BIOS:

It's a "security" software built into the BIOS of many laptops called CompuTrace. It is sorta like "LoJack" for laptops. If your laptop is stolen, CompuTrace can notify a server where your laptop is. It is written by Absolute Software and provided to laptop manufacturers so they can include it in the BIOSes they supply for their laptops.

CompuTrace is a rootkit <...> it will hijack the AUTOCHK.EXE process that normally runs during Windows boot, and instead run its own code. One issue this rootkit may cause: chkdsk will not run during boot like it is supposed to.

So, what is this thing?

I'll summarize what I've found out so far:

This thing lives in your BIOS.
When the system starts, it searches for "autochk.exe" in your system folder, supporting both FAT and NTFS drives. Then it hijacks "autochk.exe" substituting its own code instead, which unpacks and starts the "rcpnetp" process. It also verifies the registry key [HKEY_LOCAL_MACHINE\SYSTEM\ControlSet\Control\Session Manager] "BootExecute"= autocheck autochk.
When the Internet connection is up, it updates itself via the internet (connecting to 209.53.113.xxx - xxx.absolute.com), tries to send some personal info to the server and then listens to the instructions.
This crap is white-listed by most known antivirus packages that's why it was not found by my antivirus.
This crap has even created screenshots of my wife's activity and placed the JPG files into the %WinDir% folder, it gathered system reports about the laptop, our external IP-address etc. etc. etc.

How to fight this

If you find this process ("rcpnetp.exe") in the processes list, follow these steps:

Delete the files rpcnetp.exe, rpcnetp.dll from your system
Move autochk.exe to another folder
Edit the registry:
[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager]
"BootExecute"=hex(7):00,00
Note that this crap will come back after reinstalling Windows
Note that you can forget about the checkdisk tool forever.

What's the big deal?

Now, this rootkit does no harm. But

I don't like that someone collects my personal info without my permission
I don't want my computer running programs I never approved
I don't like the possibility that a dishonest employee at Absolute Software can execute remote commands on millions of laptops. There's a name for this. It's called a botnet.
I don't like the possibility that some malware authors can piggyback this system.
I want to be able to CHKDSK my hard-drive for Pete's sake!!!

UPDATE: After some more Googling I discovered that this was a big news story about a year ago, and some (not all) laptop manufacturers have released BIOS updates that remove this rootkit... But my wife's laptop was bought on Amazon only 2 months ago, it's a Toshiba Satellite T135, and the rootkit is still there. I just tried downloading the latest BIOS for our Toshiba - and the "rcpnetp" is still there even after reflashing BIOS. So check your laptops! Esp ASUS and Toshibas. UPDATE 2: Turns out that the AVG free antivirus detects this as a rootkit and tries to remove the files. But - of course - it's back up after a system restart. UPDATE 3: Here's the patent for this thing.

UPDATE 4: I've just sent a support ticket to Absolute Software requesting to remove this tool, let's wait for their reply...

Why I hate IE6. And why I miss IE6

noreply@blogger.com (Alex) — Sun, 17 Apr 2011 14:28:00 +0000

I'm getting kinda tired of cross-browser development. Yes, I know... The more the better, competition rocks, rendering standards are great, FireFox is cool, Chrome is awesome, and the evil MSIE monopoly is sacrilege.

I mean, I do hate IE6. Just like I hate IE7. I even hate IE8 a little bit. And, of course, I hate IE9 (for that lousy font rendering and no proper CSS3 support). But I do miss the IE6-days. I never thought I'd say this, but I do miss the days when IE6 was The Browser For Them All.

I still have to support IE6 and IE7. I'm sorry, I'm not some big corporation to say "OK, we're phasing out support for the older browsers". I have a freaking ton of customers running older browsers, OK? Running all possible kinds of older browsers, actually. Including the stupid IE6. I cannot just phase them out.

But even without the older browsers there are still lots of issues with the newest non-IE browsers as well. I won't bore you with the details, but trust me - both Chrome and Firefox have their nuances. And every tiny piece of JavaScript, every non-trivial HTML-code has to be tested in a zoo of browsers. Not to mention mobile devices.

Cross-browser is fine, competition is good, the more the better... But I'm just getting kinda tired of this:

Now, to make this post not so boring, here's a great "History of Web Browsers" infographic by Shahed Syed that he kindly allowed me to publish here. The history of browsers and their popularity over time:

Lessons learned from The Traffic Spike

noreply@blogger.com (Alex) — Tue, 12 Apr 2011 09:47:00 +0000

My recent blog post about the Chinese hard drive has attracted HUGE amounts of traffic. It's been featured at TechCrunch, Slashdot, Reddit, StumbleUpon and others. Of course, after being upvoted at HackerNews - my long-time personal favorite.

It was "liked" by 14K (forteen thousand) people on Facebook and retweeted more than 2.5K times. My blog has received about 450 000 visits (and still counting) - thank God I host it at blogger.com, otherwise my server would be dead by now.

But what's in it for me and my startup? Let's have a look at the ups and downs:

Pros

Self-esteem and excitement. I'd lie if I didn't say that my ego was flattered. Come on, wouldn't you be happy to get this ~~shitload~~ impressive amount of visitors and mentions?
SEO-effect. The blog has attracted hundreds of links. Those links are completely outside of my business's target keywords niche (the most weird ones came from a marijuana-growing forum, German politician's blog and Vietnamese social network)... But I guess I'll still earn some "trust-rank" for my blog.
Brand awareness. Hopefully. In Vietnam and Germany, at least.
Conversion - I gained some new subscribers and Twitter-followers (read on for the exact numbers)

Cons

Low conversion rates. And by "low" I mean "almost zero". Only 15% of the visitors actually checked out some other pages on the blog. Out of 450 000 visitors only 500 people (0.11%) have subscribed to my feed and only 100 (0.03%) have followed me on Twitter. Another 100 (0.03%) visitors cared to view the "about this blog" page and only 400 people (0.08%) have checked my main (non-blog) website so far.

I'd call this kind of traffic - "carpet bombing" (as opposed to "laser-guided missile"). Still, the SEO effect should not be underestimated.
Comment-moderation is exhausting. The Internet ~~is full of idiots~~ has some unrestrained individuals. Every five minutes some Anonymous was posting an aggressive, racist, offensive or meaningless comment. I spent hours removing hundreds of the most awful ones.
Productivity drop. You end up checking your blog every half an hour, reviewing stats or the number of retweets. And - moderating comments.

Of course this does not mean that the "viral" approach is dead. You should keep creating great "shareable" content... Just remember that the "viral" traffic converts badly, the bounce-rate is 95%. It won't drive the actual sales. But do concentrate on converting this traffic into RSS or mailing list subscribers instead.

According to this great post by Rob Walling - every traffic source has a half-life and this was a striking illustration. Traffic spikes after being featured at TechCrunch, being upvoted at Reddit, being "Stumbled Upon" or retweeted - are always followed by a decay. And your job is to keep your website prepared to "The Traffic Spike" and make the most of it.

Unfortunately, my blog was completely UNprepared. It didn't even have an "about" page until last week...

Steps to prepare your blog

Add the "sharing" widgets to your pages where appropriate. Facebook "like" button, "retweet" button and a couple of bookmarking sites - are a must. You can use a service like addthis.com or add the buttons manually.
Optimize your feed: host your feed at feedburner to track your stats, ensure fast delivery and optimized markup. Add "share" and "retweet" buttons to all your RSS feed items. If you use feedburner, go to "Optimize" - "Feed flare" and add Facebook and Declicious. The "retweet" flare is not there by default, you have to paste this line into the "Add new flare" field:
```
http://erik.thauvin.net/download/tweetflare.xml
```
Call-to-action in every post: add a "subscribe" link to the bottom of every post of your blog. This includes the RSS feed (your RSS items will get published and republished many many times).
Content is king - OK, you must be sick of this one. But, guess what, it's still true!