A Teen Programmer Built a Tool to Generate Fake Lyrics for Your Favorite Artists
genius.com
genius.com
In this case though, I'd believe them.
Partial write-up: https://habr.com/post/453232/
Took about 2 weeks to make, most on the website part. So if you want to teach your kid programming and some AI, it's not so hard.
Then you get something like this: http://billion.dev.losttech.software:2095/song/3971156712
:D
Whenever I see statements with absolute statements like anything, nothing, always, never, I question them.
We all stand on the shoulders of giants. So what we do is impressive for when we do it. What those in the past do is impressive for when they did it.
edit: yeah, figures. When calling out ElsaGate, I got replies like "your argument is literally 'think of the children'". But in this thread it's all so heartwarming, or respectively turning statements that aren't dripping with soppy general statements that mean nothing into an "attitude" or "way of looking at things" (which remains BS, and reaching 500 karma and finding the downvote button doesn't make that false, either)
The entire result was hot garbage.
However, there were buried segments that were good, but I'd have to lay that on the tightness of Mr. Miranda's lyrics as we're dealing with Markov chains.
All in all, that was a few minutes of amusement.
"Just you in the time for it's just a rabbit
Let's rock
Happiness will"
The cynic inside me thinks that people’s negative responses here are not really about corrupt media (tm). They are just reflections of the contrarian attitude that runs rampant on this site. The fact the HN guideline has to explicitly forbid 'shallow dismissal' tells you how much compulsive negativity is happening on this site.
(And yes, this post is also a display of my own negativity.)
I am not defending the media in any way. This is low quality journalism (on a site that does not seem journalistic anyway). The achievement of the 16-yo in the story is not outstanding in this field. HOWEVER, instead of reading yet another comment saying 'meh Markov chain', I guess I’ll prefer something like more advanced materials on text synthesis, some related projects, or just some advice for the author would be nice.
Credit where it's due, that's more than most 16-year-olds ever manage (though well within the power of a 16-year-old with a strong interest in programming).
"DOCTYPE HTML / if lt IE 7 / html class no js IE6 oldie!"
That polka lyric jams.
I think it's great to encourage students like Samarth and I by sharing and talking about what we're building. You can offer specific and helpful feedback which I would really appreciate. But I don't appreciate your calling my work bullshit. And I think this attitude chases away less privileged students from pursuing things like software further. There's value in making something that people get joy from, even if it's not the cutting edge shit that OpenAI blogs about. And I'm confident in what I made here.
I would bet that OP was once themselves the recipient of similar "breathless gushing" when they were young, and lost a lot of the wind in their sails when formerly impressive achievements became simply expected. Better not to depend on that kind of validation, it stops early and never comes back.
And worse, someday you might read articles like this and feel discouraged, because it seems nobody will care that you just discovered how much fun you can have with Markov chains in your 30s.
OP was actually wrong with their hueristic in this case - the article doesn't really go into your age beyond a passing mention, but that doesn't change the fact that OPs critique was with article writers, not programmers-who-have-their-work-written about.
Still feels like HN generally has a counterproductive attitude to stories of this type.
Personally, I think your project is getting coverage not at all because of your age, but because it produces interesting results that people want to see, regardless of whether the algorithm used is considered cutting edge or not.
This isn't to denigrate the work of either of the two kids involved: the program produced is interesting and fun, the encouragement is awesome, and they've lucked out with this publicity and should milk it for all it's worth.