Great blog post about l10n and i18n! I'm working on improving that process in our company and currently I'm choosing Zanata [0] as a (Java-based) translation platform because out of Transifex's no longer maintained community edition (how unfortunate!) and Pootle, Zanata's installation actually was painless and the community around it is very responsive!
Too bad I didn't stumble upon Weblate [1] first though, it looks promising (thanks onemorepassword).
I've set up an independant "localization server" that executes the following process:
1) Regularly pulls new revisions of the code and updates to the latest revision.
2) A mercurial hook [2] is thus called and the source strings are extracted from the code with xgettext [3] so that new POT gettext files are generated.
3) The POT files are finally pushed to Zanata's server via its API.
We currently do in-house translations for one locale, while others are managed by an extenal translation provider. Employees in our company can just login (Zanata provides OpenID authentication) and collaboratively translate and review the application strings. Whereas Zanata can be used to export ressource files and push projects to our external translation provider's platform via their API.
But as others have said in this thread, l10n automation curently involves a lot of manual code glueing and adapting with your version control system. There's definitely potential since available solutions only address the translation problem and haven't gone very far in the whole process.
I'd be more than glad to exchange about the subject with others who have gone through the same experience!
I think this points out the necessity to develop better proof assistant systems [1], in particular for automated proof checking [2]. However, I have never interacted with such systems and thus don't know whether it will be possible to just feed Mochizuki's formidable constructions into it.
I would not expect so. Mechanized proof systems tend to require a lot more detail than one would put into a proof meant for humans to read. There's been a lot of work in automating part of the generation of a proof, but that still requires a human to look at what the automation came up with and intervene to guide it in the right direction.
Don't forget that you would need to put not just Mochizuki's work into a proof checker, but also everything that it depends on. From what I understand, this is far too much work to be feasible with currently available proof checkers and changing that would require more than incremental improvements.
I might have not fully understood the inner workings of the classifier, but wouldn't it be interesting to allow it to learn/calibrate from user input? Basically, is there any way to apply supervised learning to your classifier?
Hi, thanks for your feedback. The classifier behind the scenes has already been trained with a supervised learning algorithm, so it can be customised/adapted to any user input. This is a business case I explore with each customer in order to properly fit the tool to the particularities of their problem (e.g., the specific salient features that represent their data).
I find that ressources on open source governance is pretty scarce on the Internet and it surprises me that there is no proper "initiative" or "think tank" to develop this idea.
Not saying it's easy, but if you want raw data, there is a lot out there.
FEC filings from the last few decades: http://www.fec.gov/data/
Bills and legislators: http://www.govtrack.us/
Lots of data, but hard to access: http://votesmart.org/
A nice tool/repository for extracting and sharing data: https://scraperwiki.com/
This is just the tip of the iceberg. I'd be happy to talk with anyone else about finding what they need.
Nice little read. In my opinion, blogging not only serves as a way to structure and express your thoughts better, it can also be an incentive to think about subjects other than one's domain.
I am a firm believer that every notion and concept, regardless of their context, can be analyzed together and combined to form new, more insightful ones. For instance - I have no examples in mind right now - there must be a way to apply aesthetic rules of gardening to team management in IT businesses!
This is where blogging and writing in general becomes essential, I think, because it comes with a responsibility to keep it somewhat alive. And this same responsibility will - consciously or not - push us to read more books about things we don't know, to undertake projects we have doubts about and to embark in new adventures, just so that we could share some of the experience with our fellow readers.
I have never been really successful at writing, and I have difficulties taking the first steps, but hopefully this year will be a good one for it!
This sure was an interesting first JavaScript project. It turned out harder than I expected, but I've learnt a lot, from JavaScript/jQuery basics to free dictionary APIs and JSONP format.
Obviously, there is still work to be done, but this is a basic functional bookmarklet. The goal will be to fix some bugs to enable it on as many websites as possible. Also, there are many ways to expand this (translation, more dictionaires, ...), but it obviously depends on available free APIs.
I'd love to hear some feedback from you (especially on speed, compatibility and user-friendliness)!
I've got a 'KeyError at /twitter/thanks' when refusing access from the app to my twitter account (no offense meant, I was just testing it out) and then clicking on "Go Back to This Year" (which, I guess, is supposed to redirect me to the app?). I can send you the full error page if you want.
Anyways, kudos on your first Python app & keep building things!
I can definitely see the potential of this, and would love to contribute. I'd go even further and suggest the ability to collaborate on side-projects, think github with broader project types. The hard (and fun!) part would be to devise a minimal set of functionnalities to power such variety of projects.
