Oh good Lord no, that would be far too easy. The API for journal downloading is the one Brad wrote in like 2000 and it's been touched like four times since then (and those were all mostly to fix compatibility problems with newer versions of Perl modules, heh). Getting data out of LJ is like emptying a lake with a bucket! It's just that everywhere else on the internet doesn't even give you the bucket, so it still is easier to interact with than like, Tumblr. (We tried to look into what Tumblr import would involve once and backed away very, very slowly without making eye contact.)
And did I mention it used to be badly documented, and is now completely undocumented because LJ deleted all the documentation? It is a miracle the importer can work at all and the only reason it does is that we knew exactly what we were working with from our time working on LJ, basically. I genuinely think anyone else trying to implement a similar feature would have broken down sobbing by the fourth time of having to completely recreate text encodings from the ground up because of a dumbfuck decision the Perl maintainers made in 1999 that caused Brad to make a "it made sense at the time even if in retrospect it looks highly dumbfuck" decision in 2000 and poison half the data in nearly every LJ created before, like, 2005. Just as an example.
The importer basically only exists at all because we are very very fucking stubborn and refuse to let the goddamn thing defeat us. We have made it work this far, goddammit, we are not losing now.
no subject
Oh good Lord no, that would be far too easy. The API for journal downloading is the one Brad wrote in like 2000 and it's been touched like four times since then (and those were all mostly to fix compatibility problems with newer versions of Perl modules, heh). Getting data out of LJ is like emptying a lake with a bucket! It's just that everywhere else on the internet doesn't even give you the bucket, so it still is easier to interact with than like, Tumblr. (We tried to look into what Tumblr import would involve once and backed away very, very slowly without making eye contact.)
And did I mention it used to be badly documented, and is now completely undocumented because LJ deleted all the documentation? It is a miracle the importer can work at all and the only reason it does is that we knew exactly what we were working with from our time working on LJ, basically. I genuinely think anyone else trying to implement a similar feature would have broken down sobbing by the fourth time of having to completely recreate text encodings from the ground up because of a dumbfuck decision the Perl maintainers made in 1999 that caused Brad to make a "it made sense at the time even if in retrospect it looks highly dumbfuck" decision in 2000 and poison half the data in nearly every LJ created before, like, 2005. Just as an example.
The importer basically only exists at all because we are very very fucking stubborn and refuse to let the goddamn thing defeat us. We have made it work this far, goddammit, we are not losing now.