I musn’t be feeling well, as I’ve signed up for both Twitter and Google Latitude in the last fortnight; surely down to a weakened sense of privacy lately.
Google Latitude on the surface sounds like a good idea, although at the moment it’s telling everybody I’m currently enjoying a town called Saint-Ghislain, in Belgium. It’d be niceI but it’s a little far from Ballina.
We were working in the (insanely expensive) Brisbane suburb of Ascot all week, and our regular Kangaroo Point house was occupied so we decided to grab one in Redcliffe, an old double storey holiday unit block called Lutanna Sands.
Since it’s right on Moreton BayI I got up every morning to catch the sunrise, but was thwarted by clouds every single morning. These are from the one morning where the sun wasn’t completely hidden.
It’s been a long, long road, but the shiny new Kiwix-powered OpenDisc is ready for download. It seems quite a while since Yann Hamon first contacted us to say that Oxford Archeology had customised the offline Wikipedia disc to run the existing OpenDisc files.
This really struck a chord with me, as the existing backend was (almost amazingly) closed source software. Not that it was intended to be, merely that over the development of TheOpenCD the source files for the backend were completely lost. It felt like a terrible cheat to be promoting OSS when the software back-end was effectively orphan freeware.I
From there Yann helped shape the foundations of our Kiwix browser, spending odd hours with myself swapping files backward and forward ironing out the littany of quirks that unfolded as we got closer. Every time it seemed to be finished we’d find something else, so luckily by this time I’d contacted Emmanuel Engelhart (developer of Kiwix) who systematically identified and solved every single problem left. OpenDisc owes a lot to both Yann and Emmanuel, as it really couldn’t have been done this well and this soon without them.
During all of this, updates for OpenDisc still had to be developed, and subsequently they were more disparate than I’d really hoped for. At times it felt a bit like the delays from TheOpenCD era, but hopefully now that’s a thing of the past. 
I’d really like to thank Yann for having the original idea and helping shape it all in the beginning; Emmanuel for offering the expert advice that only a developer can give; and not to forget the others that helped out at throughout, like David Bolton, Peter Kemp, and especially everybody who downloaded and tested our original “Kiwix Test Release” to provide feedback when we thought we were almost complete.
Patrick Patience throughout this has helped keep the forums active and held the fort when I’ve been away with work. Although I’m sure he’s as busy as the rest of us, he’s agreed to take on a new role; Patrick’s now solely responsible for the selection of new programs on OpenDisc. With our plan to rapidly expand the disc now able to proceed at full-steam, Patrick will be keeping a keen eye on both nominations and OSS in the wild for new programs to fill the enormous amount of free space on the disc,II and help steer the future of the project.
The only rough edges that we didn’t fix before release is the slightly wonky screenshot viewer, which hopefully can improve for the next version, and the barebones homepage which will hopfully be more cohesive next time around. I didn’t think this version should have waited for everything to be perfect, as most of our returning visitors I’m sure would have soon been tired of updates about Kiwix.
There really is a littany of small changes to the disc, most of which will go unnoticed – we’ve even managed to replace the old common icons with Tango Project equivalents, which really finish off the new design.
With the hurdle of Kiwix behind us, I’m looking forward to seeing where we are six and twelve months from now.
Much along the lines of PGDP and Recaptcha, The Book Oven is aiming to make social collaboration proofreading available for all publishers (public domain or otherwise), in a simple way by chopping text into sentence blocks, much in the way Recaptcha grabs single words. People discouraged by proofing whole pages on PGDP should find this more palatable, while proofing will be more accurate than the Recaptcha model as you aren’t just fixing a lone word.
In the past I’ve hosted my own photos in the shudder-inducing original GalleryI , before moving to Picasaweb, purely because I like using Picasa to manage my photos. Aside from the web interface being a bit wonky, the main problem is that nobody seems to notice your work – I got a sum total of seven comments in the 12+ months, and with no way to look at statistics I didn’t have a clue if they were being seen at all.
The catalyst for change was the Flickr app on my Blackberry, and the recommendation of a few friends that their photos really do get noticed. Before plunking down the $25 a year for a pro accountII I decided to post three sets to test the waters. One being my visit to Alcatraz (which was the most popular on Picasaweb), and two that I thought would only be of interest to a minority of people, so that I’d get a good gauge of traffic.
I really didn’t expect how well received three sets, completely unpromoted, were in the short time I’ve had them up. With a fair few comments, a request for a group addition and plenty of hits, the icing on the cake was a message from self-tour startup Schmap letting me know that one of my Alcatraz photos would be in the sixth edition of their guide to San Francisco. They might not be Lonely Planet, but it’s something that wouldn’t have happened if I were still using my old solutions. It’s a strange pick (and was taken with my tiny p&s whilst standing on a Segway) but I guess they liked it.
Hopefully soon I’ll migrate all the other photos over and succumb my twenty five bucks to Yahoo.
This month marks five years since the landing of two NASA rovers sent for the Mars Exploration Rover Mission, which is unbelievable considering they were expected to last just three months on the red planet. Rather than duplicate text, the official press release explains the history and the road ahead. There’s always the photo gallery for those less inclined to read.
In the latest issue of Popular Mechanics there’s an interesting piece of lateral thinkingI by Glenn Reynolds who really hopes that we don’t find any life on Mars. I hadn’t really thought of itII but it really does make settling and terraforming the red planet so much easier, and also gives the (albeit theoretical) possibility of returning to Earth without the risk of killing us all with foreign organisms.
In the meantime, let’s hope the Russian mission Phobos-Grunt doesn’t accidentally forward-contaminate Mars as Space.com blogger Leonard David worries about in his column.
.. is like going on a holiday; on arrival you remember something vital you’ve left behind. 
I’ve conglomerated my old domains to redirect here,I which hopefully should make it simpler. No idea what effect it’ll have on Google results, although I’m sure to learn in the coming weeks.II I couldn’t do a complete redirection because I’m using better permalinks now, and would have had to do more than 100 individual changes.
If you were pointing to one of the old domain names, please update your links because the old domains won’t be around forever. At least not pointing to me, in any case.
Edit: No sooner had I made this post an email hit my inbox, letting me know that the PGERT was not only already underway, but that there were only 1100 errors, not the 10000 as claimed before. All interested parties should make their way over to the ErrorCorrection page over at Gutenberg.org for more information, and completely ignore everything else I’ve just said.
After talking with Mike Cook and Michael Hart, clearly the easiest way to get the ball rolling on PGERT (and more importantly, keep it rolling) would be to use a simple forum system to manage and fix the errors reported.
Mike posted the ideaI of using the already populated Distributed Proofreaders forums, instead of duplicating efforts by starting a separate forum on DP-News or Gutenberg.org. That way nobody would have to wait for a specially coded solution to be made, and we could start on it almost straight away.
Broken down, here’s the simplest solution I came up with:
Open a new forum at DP for PGERT, and have four sub-forums:
The 10,000+ current errors would have to be entered into “Report an error” by volunteers (of which I’d be more than happy to help with) rather than the current plan of handing a few errors out to people interested in helping out. For these we could save time by not using the planned template, just a simple copy/paste so that we don’t spend too much time migrating the old errors over.
This would hinge on the following prerequisits:
The singular problem I can see is the reporting of new errors. At the moment they all go to the errata@ mailbox, which is how we got into the problem in the first place. The only problem with reporting errors on the forum is that people would have to join first. Sure it’d be great to inadvertantly get people to join DP, but the extra steps involved might discourage reporting the error. Ideally we’d just have a form on Gutenberg.org with say, recaptcha to prevent spam, and have some magical script format it and post it on the forum. Or have the “Report an error” sub-forum open to the public for posting.III
It’ll be simple as pie to implement, providing it can be promoted well enough and that future errors can be handled better than being dropped in a mailbox. Perhaps it might be worth nominating for the “magical script” I mentioned to be developed by somebody in next year’s Google Summer of Code.
Update: As pointed out to me, I did sound acrimonious writing this without necessarily meaning to be. More on that over at DP’s forum, and I’ve appended a footnote to the main text here.
The statistics usually found in the weekly Project Gutenberg email do get a bit tedious for me,I so I was surprised to see the latest gweekly volume (mirrored at the DP-News blog) was a nicely written piece – a call-of-arms by Michael Hart for an error correction team to be formed, and for everybody to send along the message to get the word out.
This is what irritates me; they openly call for help, but give absolutely no explanation as to the specifics. Who is co-ordinating the effort? How will errors will be reported, catalogued or fixed? Or anything else for that matter. And with two far simpler solutions, it really is potentially duplicating efforts on a project already time-poor.
Distributed Proofreaders already run Smooth Reading, where people can pick up a text and read it like a bookII and pay critical attention to sniff out errors that have slipped through the net thus far. Like every other round, SRing already has a shortage of help at hand, so why not add add a sprinkling of books from PG and actively promote it on the PG site to get the ever elusive new user into helping out by reading?
The other option would be to run a simple bug-reporting system on PG,III of which there are plenty of open source options on offer. Let people anonymously add errors (with minimal fuss, just a text box and captcha) so that the errors can be catalogued, verified and acted upon by a team of volunteers. Important to both methods is a list of false-positives, otherwise as time goes on there’ll be more time wasted on duplicates.
The key of any solution to the error reporting problem is to get new people helping out – there’s no point redirecting people already busy spending their free time on DP and PG.
An example ebook is given in the email that had a glut of unreported errors, although it isn’t specifically listed and there’s no reference that it was provided through DP. I seriously doubt we’d let 23 errors in a 300 page book get through, even though it’s less than 1% error rate. Anybody who proofs at DP would agree. One interesting sidenote is that the errors were found by somebody who was preparing a human-read audiobook, and would no doubt be looking at the text the same way a DP Smooth Reader would.
A more interesting point is that PG has received more than 10,000 emails with suggested corrections. In 37 years. Why on earth is PG still using email as the only way to report these? A lot of casual readers/visitors would no doubt notice errors and not bother to write an email, IV something a simple text box (with captcha) and a mention/link in the introductory text in the books would exponentially increase.
To be fair, the blog is a step in the right direction to building a better community around the project, although it’s hardly referenced on PG at all. They really should have user forums,V as well as open up the mailing list archives to the public that don’t want to subscribe to the actual list. The only community communications are effectively going on behind closed doors.
A one sided rant? Undoubtedly. I just hate to see time spent at DP diverted elsewhere, which deserves so much more than it receives.
[↩]
Instead of the normal channels that electronic goods pass through to get to retailers, Kogan order their stock directly from the manufacturer and sell direct to the public, cutting out the middlemen the difference is at the bottom line, with some markedly cheap prices. Not a bad effort for the 25 year old Ruslan Kogan, although some of the forums have quality concerns over the LCD displays that add to the doubt most people have about big ticket items online.
