#Debuild

[davidcreator]

Debuild: Criação de Aplicativos Web

Você está em: Início > Artigos > Inteligência Artificial > Debuild: Criação de Aplicativos Web Olá! Caro leitor, este artigo é para quem esta procurando tecnologias de inteligência artificial para ajudar nos trabalhos do dia a dia Introdução O desenvolvimento de aplicativos da web é uma parte fundamental do cenário digital atual. No entanto, muitas vezes esse processo é desafiador e consome…

[varezhka]

i really didnt want my life to be an example for future political and social scientists

[msbarrows]

My current set of starships:

Barb deBuilder, my hauler and designated construction material storage. *Smacks top of starship* I can fit so much junk in this thing!

Omnia Vincit, my living ship and designated gunship for when I’m going pirate hunting (since she has the best damage output)

Tuxedo Fairy, my explorer. I love how elegant his cream and black colouration looks.

Blind Dove, my flying wing exotic ship (white tall fin ventral wing mosquito with gold honeycomb trim). No specific role. I’d been holding out in hopes of a squid ship before spotting this one and deciding it was too pretty to pass up.

Don’t have a solar ship yet, which is one of my favourite types. Probably start looking seriously for one after I get around to maxing out the storage on my exotic.

[bandnameserver]

Debuilders

[nervespike]

bernardcapulong | MODEL NO. MGB-001 MADE IN JAPAN debuild

[hemic-hokuto]

Any unpopular opinions?

Oh boy, I do.

The Yakuza idea was dope, even at the beginning. Y’know, the monster.

The kitten method wasn’t bad, you guys are just dead okay with murder but not murdering animal unless you can eat them.

Horuda shouldn’t stay in the final game, especially if she only stays to please the fans and has no other purpose than being a decorative plant.

If the rainbow guys/gals had to be removed because they were just debuilding purpose, I wouldn’t mind.

Anime club makes more sense than gaming club since the game sets in Japan, y’all just can’t handle changes.

If some major characters had to be totally revamped, I’d be okay with that, even if I prefer the original design over the final one.

Dev isn’t lazy, you guys are just mean.

Budo is totally overrated, c’mon.

Ayando is way overrated and I still don’t understand it.

Midori is annoying, but well, that’s her purpose.

The game has way too many kindhearted characters, stop trying to make everybody nice.

The bullies aren’t a racist representation, they’re gyaru. 2018 gyarus can still look like that because there are many variation of gyaru, just like kei.

Even if they’re bad for being administration members, I still think that Mida & Muja are valid rivals because you have a good reason to get rid of them; they’re lurking at student(s).

Taeko is put on a pedestal while Taro is getting shit and that’s not fair since they’re exactly the same character.

Yankun should have the same persona than Ayano since they’re the same person.

Yandere Simulator panty heart logo is like... gross and unpleasing. Like really. Lovesick logo is dope tho.

The dark atmosphere is way too dark to see correctly for now :/ I hope it’ll be fixed soon.

[boyfae]

i got sick of trying to build my own new skyrim modlist and decided to just try and use wabbajack and now i’m just spending time debuilding a new skyrim modlist because i have to turn off all the dark souls combat mods and the ui stuff i dislike lol

[startuplevel]

Впевнений, всі хто цікавиться штучним інтелектом, вже бачили OpenAI Codex (якщо ні, ось демка https://www.youtube.com/watch?v=SGUCcjHTmGY). Але зараз хочу показати вам як ринок почав адаптовувати технологію. Шарфі працює над проєктом Debuild, який він описує як: «Describe what your app should do in plain English, then start using it within seconds». А ось його демка https://twitter.com/sharifshameem/status/1425185575645024256

[trashforthetrashgod]

Another correction, about where the blood comes from (I'm not a native speaker, so please bear with my limited vocabulary)

Over the month our bodies form thicc spongy walls so if the egg gets i fertilized it gets stuck there and has all the sweet delicious nutricions.

If we dont get pregnant the eg gets yeeted out and to show us how disappointed our body is about that it debuilds the spongy walls. Thats where the blood comes from. We also yeet out the spongy walls. That is where those jam like things come from.

Yes sometimes we not only yeet out blood but vagina sponges filled with blood that reminds you of strawberry jam.

Also since there is no egg until the middle of the cicle (abou 1-2 weeks after our period): we CAN'T get pregnant during shark week.

Can girls pee when they're on their periods?

no we just hold it in for a week.

[jpalecek-blog]

Using ccache for building debian packages with debuild

To get debuild use ccache for compiling is easy (if the underlying buildsystem honors CXX/CC variables, which isn't automatic). Just do

CXX=/usr/lib/ccache/c++ debuild

However there's another problem. Since debuild/dpkg-buildpackage sanitizes the environment, ccache won't use your normal preferences. The only preferences it will use are from /etc/ccache.conf. Moreover cache_dir has to be set in this file, because otherwise, it defaults to a directory under $HOME, but this directory is made anew for every built package (so you'll lose the cache the moment you create it).

Alternatively, if you can't change /etc/ccache.conf, set CCACHE_DIR to your cache directory:

CXX=/usr/lib/ccache/c++ debuild -eCCACHE_DIR=your_cache_dir [-b...]

Note that the order of additional arguments matter, they must come after -e.

[tastydregs]

Now for AI’s Latest Trick: Writing Computer Code

It can take years to learn how to write computer code well. SourceAI, a Paris startup, thinks programming shouldn’t be such a big deal.

The company is fine-tuning a tool that uses artificial intelligence to write code based on a short text description of what the code should do. Tell the company’s tool to “multiply two numbers given by a user,” for example, and it will whip up a dozen or so lines in Python to do just that.

SourceAI’s ambitions are a sign of a broader revolution in software development. Advances in machine learning have made it possible to automate a growing array of coding tasks, from auto-completing segments of code and fine-tuning algorithms to searching source code and locating pesky bugs.

Automating coding could change software development, but the limitations and blind spots of modern AI may introduce new problems. Machine-learning algorithms can behave unpredictably, and code generated by a machine might harbor harmful bugs unless it is scrutinized carefully.

SourceAI, and other similar programs, aim to take advantage of GPT-3, a powerful AI language program announced in May 2020 by OpenAI, a San Francisco company focused on making fundamental advances in AI. The founders of SourceAI were among the first few hundred people to get access to GPT-3. OpenAI has not released the code for GPT-3, but it lets some users access the model through an API.

GPT-3 is an enormous artificial neural network trained on huge gobs of text scraped from the web. It does not grasp the meaning of that text, but it can capture patterns in language well enough to generate articles on a given subject, summarize a piece of writing succinctly, or answer questions about the contents of documents.

“While testing the tool, we realized that it could generate code,” says Furkan Bektes, SourceAI’s founder and CEO. “That's when we had the idea to develop SourceAI.”

He wasn’t the first to notice the potential. Shortly after GPT-3 was released, one programmer showed that it could create custom web apps, including buttons, text input fields, and colors, by remixing snippets of code it had been fed. Another company, Debuild, plans to commercialize the technology.

SourceAI aims to let its users generate a wider range of programs in many different languages, thereby helping automate the creation of more software. “Developers will save time in coding, while people with no coding knowledge will also be able to develop applications,” Bektes says.

Another company, TabNine, used a previous version of OpenAI’s language model, GPT-2, which OpenAI has released, to build a tool that offers to auto-complete a line or a function when a developer starts typing.

Some software giants seem interested too. Microsoft invested $1 billion in OpenAI in 2019 and has agreed to license GPT-3. At the software giant’s Build conference in May, Sam Altman, a cofounder of OpenAI, demonstrated how GPT-3 could auto-complete code for a developer. Microsoft declined to comment on how it might use AI in its software development tools.

Brendan Dolan-Gavitt, an assistant professor in the Computer Science and Engineering Department at NYU, says language models such as GPT-3 will most likely be used to help human programmers. Other products will use the models to “identify likely bugs in your code as you write it, by looking for things that are ‘surprising’ to the language model,” he says.

Using AI to generate and analyze code can be problematic, however. In a paper posted online in March, researchers at MIT showed that an AI program trained to verify that code will run safely can be deceived by making a few careful changes, like substituting certain variables, to create a harmful program. Shashank Srikant, a PhD student involved with the work, says AI models should not be relied on too heavily. “Once these models go into production, things can get nasty pretty quickly,” he says.

“Once these models go into production, things can get nasty pretty quickly.”

 Shashank Srikant, MIT PhD student

Dolan-Gavitt, the NYU professor, says the nature of the language models being used to generate coding tools also poses problems. “I think using language models directly would probably end up producing buggy and even insecure code,” he says. “After all, they're trained on human-written code, which is very often buggy and insecure.”

[msbarrows]

Well, colour me vastly disappointed. Completed my living egg quest tonight. Was looking forward to probably killing an hour or three cycling through living ship spawns until I got one I liked.

One problem.

The new autosave feature overwrote my manual save as soon as I stepped out of my ship at the first crash site (it filled both save channels with a new save). So I couldn’t roll back to the point at which I’d made the manual save, so I couldn’t cycle through spawns. Tried a few things that I hoped would give me a new option - flying to a base to teleport elsewhere and come back, teleporting to a base in a different galaxy to see if I could do a hold-the-middle-mouse-button-to-restart-quest-locally, but none of them did anything to change it.

So I’m stuck with this living ship, whose colours and shape I dislike.

Huh... or am I? I can probably copy over the seed value or whatever from one of my previous saves, using the handy dandy save editor I have on hand.

Edited to insert: YUP, that worked! Since the seed I copied over was from my “Audrey II” living ship a couple saves back, I’ve renamed this one the Audrey III:

In only tangentially related news, a couple days ago I got tired of the placeholder solar starship I was using, and since I already had decent money saved up, went out and hunted down a small footprint hauler to replace it with. This will of course become my construction supplies storage once I have its inventory fully unlocked:

Need to think up a decent name for it. I am still disappointed that I had to say good-bye to Barb deBuilder (and all the other ships I had collected on my previous save).

Stupid building item limit :(

[threecapybarasinatrenchcoat]

There’s this app for building and debuilding habits called Habitica that works like a game: you have a health bar, an exp bar, etc. you got daily tasks that when you complete them you get more stat and exp and other prizes. you have a customizable avatar and you can buy more stuff for it, many of the options come with benefits. It’s fantasy-genre and the style is cute. You got quests with rewards and bad bosses to beat that take damage when you complete tasks. At level 10 (I think) you get to choose if you wanna be a mage, warrior, healer or rogue according to what you need most (each has different benefits). My favorite part: PETS. YOU HAVE PETS. And you can have so many of them with different colors!!!

There’s so much more about this app and I’m shit at explaining, but seriously, it’s awesome.

So my therapist has been helping me get to grips with my ADHD, and also the concept that I’m not shit at being an adult, I just can’t do things the way everyone has always told me to do them. Like every single “organize your life” books have always left me wanting to cry with frustration, and after I got hold of a copy of Organizing Solutions for People with ADHD by Susan Pinsky I realized that was because they primarily focus on “aesthetic” over “function”. And the function of most standard “organize your life books” is to “make things look Show Home Perfect”.

So the standard “hide all your unsightly things by doing xyz” may look nice for the first week or so, but by the end of the week it’ll look like a tornado made of pure inhuman frustration ripped through the house as I try to find the fucking advil.

To give you an example of the kind of hell I’ve been fumbling my way through the last 20 odd years: dishes will be washed and left in the drying wrack but never put away. Which means I can’t wash more dishes, which means dishes pile up, which means I can’t make food, which means I don’t eat, which means my CFS gets worse, which means I don’t have the energy to put the dishes away, and so on so forth until I have a meltdown, cry to ETD (who also likely has ADHD but has never had it confirmed) about how I can’t cope with life, and then we fix it for a while, but inevitably end up back at square one within about a week.

Pinsky’s solution to this was ��remove an obstacle between you and your goal, if that means taking all the doors off your kitchen cabinets to make things easier, so be it.”

And lemme tell you, fucking revolutionary.

Laundry never ends up in the hamper??? why???? is it a closed hamper??? Remove the lid. Throw it out the window. Clothes are now miraculously finding their way into the hamper??? Rejoice????

Mail ends up spread out over every available flat surface? Put a sorting station right where your mail arrives. Put a shredder or “junk” basket under it. Shred or dump the junk immediately. Realize you only actually have two real letters that need attention, feel less overwhelmed, pay your bills on time.

Like I’m not saying this book is miraculous, but it did help me realize that I was effectively torturing myself by trying to conform to certain ideals of “perfect house keeping”, and presenting a certain image rather than just allowing myself to live in my space as effectively as possible. And why? Why was I doing that? Cause people with different lives and capabilities are perceived as the norm? Fuck that. If this was a physical problem I wouldn’t be forcing myself to conform to an ableist standard, so why am I doing it with this?

My lived space will never look a certain way, and that’s okay. It will never look show home perfect, and that’s okay. It will likely always be cluttered and eclectic where nothing matches, and that’s okay. Sometimes I will have odd socks on because sorting them out required too much mental energy, and that’s okay. Actually fuck sorting socks, just buy all your socks in the same color. Problem solved. Boring sure, but also one less thing to do, which means more time to hyper fixate on fun things. Which really, what else is my life for if not to write screeds and screeds of vampire shit posts, I ask you.

[jaggannattsaggvekar]

A developer outdated a instrument from the AI firm Elon Musk cofounded to fetch an app that means that you simply can present web sites merely by describing how they work

A developer outdated a instrument from the AI firm Elon Musk cofounded to fetch an app that means that you simply can present web sites merely by describing how they work

<img alt="Debuild Sharif Shameem GPT-3" data-content-type="image/png" data-first-image="" data-srcs="{"https://i.insider.com/5f4973d960bf1f0028079541":{"contentType":"image/png","aspectRatioW":4,"aspectRatioH":3}}" src="data:image/svg xml,“>

Debuild cofounder and CEO Sharif Shameem outdated GPT-3 to present a program that permits customers to present a web set of…

[mugasofer]

Oh sorry, I thought you meant GPT-3 in general!

Yeah, this "debuild" thing could easily be a scam, although GPT-3 does seem to be able to write code ... they keep mentioning that their current version doesn't work every time & what they've shown is just examples, which makes me wonder if they're iterating multiple times on prompts by hand when they get requests through the form to pick the best ones. It would be easy for them to justify as not really faking.

I find GPT-3 pretty worrying!

You can describe websites to it in natural language and have it code them!

(I’m a little skeptical of this demo, but not very skeptical.)

It’s probably easy to make this architecture much more powerful simply by training a bigger version…

2 years ago, GPT-1 was interestingly useful pretraining and adorable with its “sentiment neuron”. 1 year ago, GPT-2 was impressive with its excellent text generation & finetuning capabilities. This year, GPT-3 is scary because it’s a magnificently obsolete architecture from early 2018, which is small & shallow compared to what’s possible, with a simple uniform architecture trained in the dumbest way possible (unidirectional prediction of next text token) on a single impoverished modality (random Internet HTML text dumps) on tiny data (fits on a laptop), sampled in a dumb way, and yet, the first version already manifests crazy runtime meta-learning—and the scaling curves still are not bending!

… and we’ll probably find out within a few months exactly what that means!

GPT-3 has been estimated to cost $5m in compute to train, and - looking at the author list and OpenAI’s overall size - maybe another $10m in labour.

Google, Amazon and Microsoft each spend about $20bn/year on R&D and another $20bn each on capital expenditure. Very roughly, it totals to $100bn/year. Against this budget, dropping $1bn or more on scaling GPT up by another factor of 100x is entirely plausible right now.

…

I think we right now have the ability to build an orders-of-magnitude more powerful system than we already have, and I think GPT-3 is the trigger for 100x larger projects at Google, Facebook and the like, with timelines measured in months.