… or there again, there might not!

This is just a heads up for a couple of events/workshops that I’m involved in over the next couple of weeks and months.

Firstly, next Tuesday is the AGI Northern Group Where2.0Now one-day conference, at GeoPlan in Harrogate. If you want to know what this whole “neogeography” thing is, and what it means to you, then be there or be terribly antiquated. There are some great speakers lined up (and me, but beggars can’t be choosers), and it’s looking like a good day. With luck and a fair wind I will have “cool things” to show too…

Secondly, in January 2010 I’m helping on a 2-day workshop at Lancaster University on open source GIS. We did this last year, and it was well received, so it’s getting a reprisal. There’s a flyer here, and you can book here. For UK higher ed or other educational types, it’s pretty cheap if you ask me, and the food’s good too.

Hope you can make it to one or both of these.

I also did a talk last week to local government types, on how open source GIS could be viable within their organisation. The slides are up on slideshare if you’re interested!

Apologies to the couple of people who were kind enough to report portable GIS bugs on the launchpad site, only to have their bugs totally ignored.  I didn’t set things up properly, and wasn’t getting email notifications. I’ll deal with the issues and post a fix if I can, and now I am getting notifications, so if anyone else finds any issues do let me know. I also don’t know how I managed to lose my contact form, but it’s back now. Must have been pixies…

To keynote or not to keynote… I chose not, so missed out on the triumvirate of ESRI, Ordnance Survey and Pitney Bowes and instead watched a series of talks ostensibly on “the GeoWeb” instead.  By the time Andy Allen from Cloudmade finished his talk I felt like I’d been run over by an unstoppable OpenStreetMap juggernaut (in a nice way, you understand). I had a bit of an epiphany about their flexible data paradigm, after all, how could you tag a road in the West Bank as one-way if you’re Palestinian and two-way if you’re Israeli without it? More “Open” from John McKerrell from mapme.at, talking about the OSM alternative to Google StreetView, imaginatively entitled “OpenStreetView“. It’s at an early stage but promises a lot, and they are addressing privacy concerns quite nicely.

Martin Daly of CadCorp won the award for the most interesting title (Human sacrifice, dogs and cats living together, mass hysteria), and of course gets points for showing the actual clip from Ghostbusters where that quote comes from. The main thing I took away from his talk about neo and palaeo was that it’s all still geography regardless of what label you put on it, and that it should be about what’s good, not what’s new.

Brian Norman from Earthware did a great talk about creating applications for Real Estate and Travel, hampered only by the fact that he had to do a live silverlight plugin installation. I hadn’t really thought explicitly about the way Estate Agents would want to censor mapping data (showing you the nice park nearby, but not the nightclub). I also hadn’t considered their need for more detailed, up to date imagery to ensure that, as the visitor, you’re not put off by out of date pictures of half-built extensions, or the dreaded grey box telling you to zoom out.

Winner of the best presentation, as voted by the punters, was the BBC with their Story-telling on Maps. It’s amazing what you can do with the might of the BBC R and D department, and lots of help from the Ordnance Survey! To be fair, what they have produced is a very slick API for tying movement on the map to actions in a video, and it’s incredibly well presented. There was a collective gasp from the audience when they rotated a piece of raster mapping, and the text stayed at the correct rotation… (a gift from the OS and not something us mere mortals can do).

I thought it was a little unfair that the afternoon’s sessions from Ed Parsons and Peter ter Haar were changed on the hoof from simple back to back presentations to some sort of boxing match. Ed got to deliver the presentation he had prepared, whereas Peter had to ad lib responses whilst trying to give his own talk. Having said that, Ed’s demonstration of the idiocy of derived data was an absolute masterpiece and Peter didn’t stand much of a chance. This is a shame as he was trying to launch some fairly innovative (for the OS) new products including (finally) OS on Demand- a service based delivery system for data.

The concluding plenary put a lot of the previous presentations to shame. 15 and 17 year olds from Leeds Grammar School, along with two of their teachers, presented on the use of GIS within all aspects of their curriculum, not just geography. It really was great to see GIS being used so innovatively, and though there was some unease on the twitter back channel about the ESRI influence, that shouldn’t detract from their achievements.

On to the concluding remarks and prizes.  Steven Feldman stepped down as conference chair, and seemed genuinely sorry to go. Everyone, in fact, seemed genuinely sorry to see the end of the conference. I think the organising team got the “community” aspect just right this time round, even more so than last year.

My own take on the trends from this year- OSM, all over the place, and in particular Walking Papers. The neo/palaeo debate, even amongst people who claimed not to care. Frustration about Ordnance Survey derived data and licensing. Twitter as a valid conference tool. All things beginning with geo. Roll on next year…

The AGI conference last week in Stratford-upon-Avon was well worth attending, with (I thought) a really good vibe and some great presentations. I thought the twitter feed, new for this year, was a real hit, as was the ability to see talks online via slideshare soon after they had been given. The twitter feed in particular gave you a chance to see what other people watching the same presentation were thinking, and occasionally caused some jealousy as people realised they’d picked the less interesting track!

Steven Feldman’s final introduction as chairman of the conference is probably a good place to start for a feel for how it went. Attendance was up from last year (600+), which was reassuring, given the financial circumstances, with a more international spread of attendees- great for a predominantly UK-based conference. He said it was no longer about “how” you did something, in other words using packages X and Y, but “why”.

The conference tagline was “Realising the Value of Place”, which is quite clever and multi-faceted. “Place” is not the same as “location”. It’s about asking why people feel happier in one place than another, and why life-expectancy differs between London Boroughs. “Value” can also be taken in a number of ways. There’s the value of a place, mentioned above, but also as an industry in a recession we need to learn how to get financial value from what we do, and controversially, how to get value from “free” (Steven’s term, not mine), as it’s not going to go away (Yay).

The two keynotes, from Peter Batty and Andrew Turner were also interesting. Peter described the current climate as a geospatial revolution, as the industry migrates from the more established mainstream technologies such as desktop GIS to more disruptive technology such as the web and crowd-sourcing. This was the first mention of OpenStreetmap, and in particular Walking Papers, but believe me it wasn’t the last…

Andrew Turner stirred the Neo/Palaeo pot (again not the last time this came up), but perhaps came closest to defining the difference between the two- as a shift from tool-centric to user-centric. Actually this ties in very well with Steven’s comments about moving from the “how” to the “why”, and also with Peter’s comments about disruptive technologies. I think the one thing that’s very clear is that it is a total mind-set shift, and people (or organisations) that don’t adapt or evolve will be become irrelevant. Someone asked the question “how do we make money from this?”, and again there is a total shift here. Massive license fees simply won’t work in a market where people know about crowd-sourcing, free data and micro-payments a la iPhone apps.

Surprisingly, the best paper I saw in the two days, and a deserved winner of the committees best paper, and a runner up for the attendee’s best presentation, was Robert Barr’s talk on Core Reference Geographies (CRG). I didn’t even know such things existed till then, though logically they should. From the UK’s Location Strategy these are: “Commonly used geographic datasets that provide a framework for linking and integrating other geo-referenced information as well as providing key contextual information”.The establishment of CRG in the UK have been talked about for several years, but only ever talked about, yet they should be absolutely fundamental. There needs to be a cost/benefit study for creating these CRG and making them available, and also an analysis of what it costs not to do it. Robert made the comparison between the CRG and other Core Reference datasets such as DNS. The same sort of funding method (pay for inclusion but not for use) could potentially be used to fund the CRG. The one negative point I had was the lack of reference to the spatial data themes talked about in the INSPIRE directive, as it seems to make sense to ensure that these (if mandated) are all core datasets.

Another stand-out presentation on Day One was on the historical development of “place” by Martin Laker. He talked about how current boundaries in fact have a heritage going back to the Black Death, and even earlier. Clearly the geography of the UK has always been tangled up and complicated (cf with the difficulty in setting up the CRGs), so all government has to do now is blame it on the Plague…

James Cutler from emapsite presented on the Geoweb’s cultural heritage (sorry, can’t find the link), but I got frustrated when he basically dismissed the problem of data licensing by saying that it’s not really all that expensive. It became clear to me that archaeology, and perhaps other environmental disciplines, have a use-case that is totally under-represented in the great licensing debate.

Day One concluded with the GeoCommunity Soapbox, a new invention for this conference. Speakers were given 5 minutes and 15 equally spaced slides, to talk about anything “geo” that they wanted. When coupled with a live view of the twitter feed and free geobeer this was a recipe for carnage and I think it’s probably good that the wifi (and hence the twitter feed) collapsed under the strain early in the proceedings. The best soapbox rant was definitely Ian Painter’s, now a veritable internet sensation.

General trends- lots of Neo/Palaeo discussion, despite exhortations that “I’m not Neo/Palaeo (delete as appropriate) but…”. This mind-shift clearly worries a lot of people, and the industry is in a process of change as it tries to re-position itself. OpenStreetMap and allied projects are definitely on the up. The back-channels (twitter in particular) were just as important as the presentations and the face-to-face discussions.

Day Two to follow…

Just a quick note to say that I’ve been away on holiday for a fortnight, in gorgeous Orkney in the far north of Scotland. A fortnight of absolutely no computers (apart from downloading digital photos), wandering around beautiful islands with sandy beaches (OK, mostly in the driving wind or pouring rain), visiting Chambered Cairns, drinking whiskey and generally chilling out. I have to say that I very much enjoyed disengaging from technology, information streams and general online interaction very much, so obviously needed the break! I’d post a photo or two but haven’t got round to QA-ing them all yet!

I’m off to the AGI conference in Stratford this afternoon, and would welcome the opportunity to meet up with folk while I’m there- we’re intending some kind of informal OSGeo UK meetup on Thursday but I’ll be around for both days. I’ll blog about the conference while I’m there if I get the chance.

As someone else said recently, the advantage of catching up on several weeks of RSS posts all at once is that you see some trends and relationships that you’d probably miss otherwise. One that caught my eye was this, from Martin Daly, in response to a long and thought-provoking piece on open source by Ian Bicking. Without trying to second-guess either Ian or Martin, it’s clear that there are always going to be different motivations for adopting and working with open source. Via a tortuous chain of links I revisited this post of Paul Ramsey’s from last year, responding to a Jack Dangermond interview, in which open source is mentioned and summarily dismissed. Paul is uneasy with the political connotations of calling open source a “movement”, but for some people that’s clearly what it is.

I’m beginning to see open source as being a choice similar to choosing organic food, or going green. For some people, this is a political movement. For others, nothing else makes any logical sense. For others, it’s a purely market-driven decision, and I’m sure there are many more motivations. The different camps don’t always sit nicely together, and occasionally see each other as harming the general cause. But we should all take heart from the fact that going green used to be the province of the yoghurt-eating, hemp-wearing hippies, but we’re all recycling and changing our light-bulbs to energy savers (and even eating yoghurt and wearing hemp) now.

I’m writing a short paper on what the open source geospatial space is going to be like in 5 years time. I’ve got some ideas of my own, but it seems apt (and would be mighty helpful) to seek advice/views/opinion from the community on this point.  I’m particularly interested in the emerging trends that people see, and the impact that they will have on the acceptance and use of open source geospatial software in the more general geospatial “industry”.

Comment below, or use the contact form. Anything that I use I’ll credit.

Over a year after releasing version 1 of Portable GIS, and over six months since I started planning a new version, I’m pleased to announce the release of version 2! I’ve set up a Launchpad site to track bugs and answer questions, and there’s also a google group. Click here to download it, and if the download server creaks under the strain, please try again later!

The new version comes with a self-contained installer, the most up to date stable versions of all the constituent packages, a control panel, and much improved documentation. It requires 1.3GB of space to install in.

I am also pleased to announce that this release comes with the option of enterprise-level support, customisation and deployment, via provided by OA Digital. Please see this link for further details of this exciting new service.

Thanks to the Linfiniti Geo Blog, we get what has to be the most unintentionally hilarious article ever, about Oracle Xe.

It’s an open source blogger’s dream post (all quotations are direct from the article). We get fear-mongering about open source “maintenance, support, and security headaches”. We get limitations built in, ostensibly to make it “easy to install”. We get accidental admissions that “if you can reduce your EE license costs by even a single CPU, you’ve made your effort worthwhile”, and the crazy notion that we should “reduce the load on enterprise hardware” by installing databases on desktops instead. We do, however, get “A New Type of Support: The Community”, but beware, because “you won’t be able to create a Technical Assistance Request (TAR) for XE issues regardless of the support contract you have”.

It’s worth checking out some other posts by the same author, in particular this comparison between Oracle 10, PostgreSQL and MySQL, where he admitted that he was “strongly biased towards Oracle and fully expected no real competition”, and then found that PostgreSQL came out tops in his tests. Oh well, props to him for publishing it, I guess!

OK, OK, these articles were written in 2006 and 2005 respectively, and while pointing out the unintentional absurdity of this mindset is fun, it’s not big and it’s not clever. There are serious points to take away here. This article unintentionally highlights all the reasons why you should avoid the proprietary software model, like the need to put limitations in the products you want to give away for free so you can justify selling the fully featured versions at a much higher price. At the same time that it implies all sorts of bad things will happen with open source, it talks about a community for support (though not for real technical support, because that’s one of the limitations of their free product).  We (in the open source community) can see that this model is broken, and that we have better products, better support, better community. As has been said elsewhere, what happens when everyone else wises up to this?

Recently I watched a video by Seth Godin that talked about how things are broken, which, as well as being really funny, made me realise that so many things are indeed broken. Take the firefox extension update process for example. I appreciate it letting me know extensions are available automatically, but when it finishes it leaves you on a window with a single button saying “continue”. Why? If that’s the only option, then why not simply “continue” without my input? If you’re expecting me to make a choice at that point then I need more buttons…

Anyhow, in case anyone was in any doubt that licensing of proprietary software is broken, then this is a great example of the sheer absurdity of it all. For those who don’t want to click on the link- once you license the software in one country, you can only use it in that language, whether you want to or not. Yep, definitely broken.

Crikey,  that’s the first time I’ve left it nearly a month between posts! At the moment it feels a little like one of those games you play when you’re a kid, and someone shouts “red” so you go and hit the red post, then they shout “blue” and you dash to blue, then they shout “red” again, then “green” really quick before you’ can catch your breath, and before you know it you’re stuck in the middle unable to move. So what, this happens to everyone, I know, I’m not looking for sympathy.

Many exciting things have been happening though! I’ve kept quiet here about our technical consultancy, OA Digital (though if you’ve seen me speak over the last year you’ll have heard about it), but we’ve been helping to provide open source advocacy services to the Welsh Assembly Government, at the invitation of Environment Systems and One Bright Space, and we’re working on some interesting web-mapping projects involving Roman Kilns, based on PostgreSQL, FeatureServer and OpenLayers, and Planning Applications, using MapGuide Open Source.

That doesn’t leave much time for anything else, but I am also at the final stages of readying Portable GIS version 2 for general release. It has been ready for a while, but I sent it out to some testers, then what do you know, PostgreSQL 8.4 and PostGIS 1.4 came out, so I’ve got some updating to do. The new version has a much swankier interface and a proper installer (with a 500MB exe to download rather than a 1GB zip file- that’s progress).

Finally, for “fun” and in my “spare time”, I’m also working through the GeoBI offerings and Geoserver/PostGIS versioning. There are some good posts brewing on those just as soon as I can find the extra hours in the day…