dustrider
As kids we used to have great fun knocking rocks together around sunset to get them to call back. Kinda like beetle bird calls.
My view mostly it was a confluence of poor dev experience and over-engineering that killed them.
Some of those protocols were well designed. Some were secure, all were pretty awful to implement.
It’s worthwhile calling out REST as a long term success. Mainly because it was simple and flexible.
Whether MCP will have that staying power I dunno, personally I think it still has some flaws, and the implementation quality is all over the shop. Some of the things that make it easy (studio) also create its biggest flaws.
The problem for me is that it’s not worth running these myself, yeah I may pay attention to which model is better at tool calling. But what matters is how well it does at my use case.
In my team I see examples like yours, but I also see engineers having to clean up slop when one of them got over their skis.
My _belief_ is that it’s a net positive, but we’re far from taking that as a hand wave fact. And it certainly ain’t the 5-10x people are shouting across the board.
See the MIT Nanda study, and the other one from a few weeks back on the perceived vs actual productivity increases.
There is value, but so far nowhere near as much as the people pushing AI would like you to believe has actually been delivered
Unless it’s proven to be a murder, they can’t report it as such. They can say an investigation is launched etc. they’re pretty good at following up with subsequent articles or even changing the headline once facts materialise.
I agree with your base point though. “Demand better” should be the new war cry
Vote with your feet and your wallets.
But I agree at least there’s a chance to do something worthwhile.
Performance exceeded what was available at the time, scaled super well, but not something I’d ever go back to.
- xslt is not a programming language. As soon as you start using xslt:functions you’re doomed.
- it is not a fit for interactive UIs. It’s good for static content, trying to mix in js goes wrong very quickly.
My opinion is that these days infra is cheap enough that you get all the same benefits by doing standard templating in your language of choice from your data in json or whatever. Which I know is essentially the same thing without the specialist software and dsl. And that’s kind of the point.
> Limit our brain diet the same way we try not to eat (delicious) fatty food every day to avoid the total collapse of our bodies.
Gives me hope that after our societal binge on social media and genai, we will come up with healthy patterns.
It only took us a hundred years or so after the Industrial Revolution to start adapting habits after the relative plenty that resulted.
I can see llms going much the same way.
IMO There’s a market expectation that “big tech must make cuts” and therefore they either feel their share price would be under pressure if they don’t show something, or more cynically they see an opportunity to trim a little that would look a lot less worse than if they did it another time.
It’s easier to manage when all you do is assign work, reward with money and blame silent quiting, millennials, or whatever the latest cause of people “just not working enough”.
As opposed to actually caring about your staff, making sure the work is fulfilling and that the team are bought into what they are doing at an individual level.
Leading shouldn’t be about what makes it easier for the manager.
Even while it won’t break anything you want CI to be your safety net, flagging a local setup as being wrong is more valuable than magically autocorrecting it.
From my take, an interesting way of dealing with scalable workflows and batch jobs. But I need to read more into it.
I also don't think that it's a case of increasing impressions. I do however think that it allows them to display more ads, and therefore increases the chances of an ad being relevant, and as a result more clicks.
Because you're now getting accustomed to the results changing as you type, having different ads appear as you complete your query is not more obtrusive than the core function.
As an example, searching for employment tribunal brought up about 5-6 different ads (just in the main results area, not counting the sidebar) whereas without instant, their inventory is limited to three.
I would say that where there are multiple options to a character sequence it gives them better value per ad as well as they can pick the ads that are worth more, until of course the character sequence excludes them.
So in the end, you've got increased inventory and better value per ad for Google, increased speed and responses to users. Win-win
Though there are more than two parties in the search ecosystem, the tail-optimised sites are going to be left out in the cold, not necessarily the small ones, but anything that relies on 2nd page or below the fold traffic.
This may or may not be a good thing for the internet, but it's going to hurt a lot of companies.
By the way, not sure if anyone else has noticed this, but it's got a definite sense of history enabled as well, with subsequent queries for the same terms being much quicker to latch onto the directions chose in previous instances.
Giving the user access to stuff they want to buy, when they want to buy it and where they want to buy it, without giving them anything they don't want, when they don't want it, etc is a great value offering for both user and advertiser, it's great that you've got the attitude that advertising has to be beneficial for the user, but remember that it's also beneficial for the advertiser as they don't have to waste money or time on people that wouldn't buy.
If you've got access to your users bike choices, mileage and location, just contact all the bike servicing shops you can manage and offer them a deal, I will bet money the majority will take you up on it.
Those two areas, possibly along with linguistics (data mining again), were much more useful to me than the programming 101 type courses.
That said I still greatly value a CS degree (even a self-study one) simply because of the fundamentals it teaches. It's not an automatic indicator of ability, but having that level of understanding makes the difference between someone that can think conceptually about and around a problem and someone that has a set number of tools to solve problems with.
I don't think that getting a CS degree is the only way to get that kind of understanding, but I do think it's one of the easier ways. If you're a hacker on your own and you've got to trawl through the glut of bad coding books out there, it's tough to get a good education.
Though I firmly believe that if you're the kind of guy that has gotten that education without any help then you're probably better than the guy that got it with help.
I'm firmly in the camp that holds that an education is to educate you and provide you with skills and tools to operate effectively, not to obtain a certificate. Holding the certificate is not an indication that you have actually received an education, the only indication is your ability, ambition and results.
For what it's worth, I think the majority of the books on the list are must-reads for anyone in programming, there may be a few better ones (I'm a Tannenbaum fan when it comes to OS's as an example) and there may be areas that aren't covered, e.g. usability and CHI (Design of everyday things would be my suggestion there) and there is suspicious lack of anything web, mobile or distributed, but if someone internalises these, they'll definitely be better for it.
or this one which is slightly harder: http://linux.thai.net/~thep/datrie/datrie.html
Or, if python is your bag, there's a scraping lib called scrapy that was opensourced last year that's ok.
On ScraperWiki, they need source control as the environment is actually run on their servers, I'd guess a bespin type implementation, so you put all your code on the wiki, and it can get augmented etc. hence needs SCM. At least, for the moment, it seems they've got plans of releasing their api engine or at least calls to it in which case you'd end up doing the code locally, and can use your own SCM.
Their confusing naming could be explained that their not pitching this at coders, but rather journalists to try and get them to use the vast data resources out on the net.
We're actually talking to them about hosting one in South London in the next month or two.
Not sure what their plans in the US are, but it could be worth dropping the idea to some fellow journo's and seeing if there's an oganisation willing to act as host.
EDIT: forgot to mention, if you're interested in data you might also want to take a look at the open data initiative (data.gov.uk) and the guardian's data store http://www.guardian.co.uk/data-store apologies for the UK focus.
With FUD and spin making such big motions on startups (see diaspora 2 months ago) you're as well off making random bets, bets on personality or some other arbitrary factor that works for you.
There isn't a winning strategy, or all invested startups would succeed, so it's pretty much a die roll. People like PG pick a few criteria that work for them and to some measure stack the deck in their favour, but it is and will always be a gamble.
favorite: "But use it exactly how I tell you to use it, because fuck you, it's my code. I'll decide who's the goddamn grown-up around here."