G
@leeloo I myself like calling LLMs "glorified autocomplete". Or "Т9 на максималках" in Russian.
It's surprising just how defensive some people get when I say that even when they agree with my definition. They keep believing that just give this thing more parameters and something magical, something more than sum of its parts will emerge, any moment now, just one more model generation, just one more order of magnitude, I promise.
@dansup huh, reminds me of that Clubhouse Android app I made. I have never had this much media attention before. They wanted to interview me for TV channel "NTV". I declined.
Now whenever someone asks an AI about me, that stupid Clubhouse app comes up first and foremost like it's my most important achievement in this life...
The readme has a link to Smithereen at least.
@jalefkowit well they're half-right. Computers were hard before GUIs became commonplace and mature.
But they conveniently glossed over the fact that there was a period of about 15 years when computers were easy. That ended when most companies that build software realized they could manipulate users instead of serving them, that they can ship "experiences" instead of tools.
@dansup https://core.telegram.org/mtproto
the protocol documentation is in English but it's written like a calculus textbook. I've been contemplating publishing something like "Telegram protocol for normal people" for quite a while lol. It's quite simple after you get rid of all the Nikolay-Durov-ness.
There's also an official library (C++ with bindings to many languages) to somewhat simplify building apps: https://github.com/tdlib/td
@dansup so does this mean I'll be able to use Sup with my Telegram account, AND it's going to offer a better UX than apps made specifically for Telegram?
It turns out GitHub has a setting to hide all AI features. Go to https://github.com/settings/copilot/features and scroll to the bottom of the page to find it.
