ZoomerCretin

Yep. My second Dell laptop had the screen hinge break, because they anchor the damned things in plastic, and the screen case is rarely rigid. Google "Dell laptop hinge issue" and see how many results you get, and how far back they go. At least 15 years.

There's no reasonable explanation aside from planned obsolescence.

> but it doesn't logically follow that similar housing couldn't have been built without the parking rules change, just that it's cheaper to build without it.

Cost is a very real and important constraint. You can't just throw more money at a problem if the problem is that you cannot profitably provide a good at a price that is affordable. You cannot change the price the market will bear, but you can cut your costs by lobbying your city government to cut parking minimums.

Having an exception process is ripe for abuse. We have laws for a reason: it is generally a good thing for the same rules to apply to everyone. When you require an exception, you can easily threaten and extort whatever you want, because the "default" is to deny an exception to a rule. We see this all the time with planning: planning commissions extract expensive "concessions" from developers all the time. "Oh, you technically can build that much, but it won't be profitable with all that parking we mandate. We'll gladly grant an exception, if only you capped heights, and have fewer studios."

Bad judgement? Sam Altman is a prolific liar who attempted to oust a board member by spreading different lies to different people. He's not even an engineer! He has established a cult of personality and popularity, and that's it. They were absolutely right to try to oust him. The only mistake was in doing so in such a ham-fisted manner.

If it's trivial but nonfree, it still runs. What is meant by non-trivial, and how do you programmatically determine that a script is non-trivial? Like this: https://git.savannah.gnu.org/cgit/librejs.git/tree/common/ch...

* If an identifier on the banned list exists, it is non-trivial. The list is here: https://git.savannah.gnu.org/cgit/librejs.git/tree/common/fn...

* If an identifier is ever followed by the square bracket indexing operator, it's non-trivial. (Looking at Stallman's requirement, it looks like this should only apply to accessing properties on an object, but the authors don't seem to have found a way to differentiate using [] on an object's identifier and [] on an array's identifier)

* If there are more than three loops, it's non-trivial.

* If it uses fetch(), chrome, browser, XMLHttpRequest, or eval(), it's non-trivial.

Oddly, referencing "document" is commented out, even though Stallman explicitly said any DOM modification makes a script non-trivial. I wonder what broke (outside of what they expected to break).

Tools like this seem like a great way to manage breaking changes if done right. Far too often, frequently with JavaScript, I'll find myself with a project whose dependencies are 2-3 major versions out of date, and the documentation of these versions' breaking changes is of course very poor and lacking in sufficiently detailed examples.

Yes, but since the US is becoming extremely protectionist/nationalist with its energy infrastructure, the only relevant bit from this story is that this production is occuring entirely within the US.

Though it is good news that this company can produce much denser batteries in the future!

From the article, these are estimated at 70 Wh/kg. Pretty abysmal for EVs, but the trade-offs likely aren't too undesirable for home battery backups or utility-scale batteries.

Yep. The industry is not the same everywhere. The high paying jobs on the US coasts hire for intelligence and general programming skills. Everywhere else, you're expected to have specific experience in their language and framework. Every job I've had, I've overheard applicants being dismissed out of hand because they didn't use the exact language we were using. It's ridiculous, but it's reality.

If you can learn it quickly on the job, you can learn it before you apply for the job.

> Vacant land attracts higher taxes in Japan than land with buildings

A hell of an incentive to build or sell your land to someone who will build. We've struggled with land speculation in the US for longer than we've been a country despite an obvious solution being available to fix this. We really could benefit from taxing land value alone, regardless of the value of any structures on top of it.

There was no bipartisan consensus until Israel's atrocities began last year.

Israel, who is mad that there exists a major social media app which does not censor pro-Palestinian content.

Poor drivers who are likely to cause accidents should pay higher rates. If you don't like it, you're more than welcome to drive a 10 year old clunker.

There is no difference between what GM did and what any social media company does. At least with GM selling driving data, there is significant public benefit to making bad driver pay for their poor driving instead of the rest of us having to pay for it.

Both major presidential candidates are very anti-China. What exactly would they gain from election interference?

So this is almost certainly to ensure it lands in the hands of pro-Israeli censors.

Yep. TikTok is the only major social media app in the US that does not downrank pro-Palestinian content. No one in the US government cared about TikTok until the latest conflict in Israel. This is the only reason for the ban.

Wow. Another incredible bar raiser for open-weight language models. I can't wait to see a fine-tuned version.

Yep. Google has TPUs, and AMD/Nvidia/Intel/Apple/Qualcomm all have tensor coprocessors now. From a CPU or GPU to an FPGA, the cost/benefit is huge. With every device having tensor cores, not so much. ASIP and ASICs are likely the way to go, at least for common operations like matrix multiplication.

I'm glad to see this is being studied. I did a brief half-semester project summarizing the usefulness of accelerators after the death of Dennard's law and Moore's law (this was when Intel was still pushing out incremental improvements on 14nm). The short summary is that accelerators offered substantial performance per watt improvement, at the cost of longer development timelines, cost of manufacturing in the case of ASICs and ASIPs, and the inability to fit large algorithms on FPGAs. Though Microsoft found a way to daisy-chain FPGAs for Bing's page rank, it just didn't produce the order-of-magnitude improvement necessary to justify moving to FPGAs. With smaller algorithms or bigger FPGAs or occasional operation offloading (like what we do now with TPUs and GPUs), they could be good candidates for accelerators.

Of course the real killer is how easy it is(n't) to develop for them. AMD offers much better performance per dollar than Nvidia, but of course their poor drivers make using their hardware a fool's errand.

RADs sound like a good idea. We should be making reprogrammable hardware available and easy to use for everyone.

As opposed to the private industry, famously meritocratic and never hiring friends, family, and insiders?

The benchmark for the latest checkpoint is pretty good: https://x.com/teknium1/status/1780991928726905050?s=46

This opens the door wide enough for extortion and the end of net neutrality, and for this to happen again:

> Second, that the company repeatedly promised reliable, “no buffering,” “no lag” internet, especially to services like Netflix or to online games like League of Legends, but was in fact purposefully letting the interconnections between TWC and outside companies degrade to an alarming degree — unless the companies, like Netflix, were willing to start paying for access to TWC customers.

> https://nymag.com/intelligencer/2017/02/time-warner-cable-la...

So rationing? I wouldn't call that fantastic. Why should someone who's barely interested but willing to go for $10 more deserving of a seat than a super fan willing to save up to shell out $150 for a seat?

had

Wow thank you for the insight. I appreciate you taking the time to write all of this out, and also for your stubbornness in testing!

That statement has two qualifications.

Yes that's why I qualified "serious" dogfooding. Of course you use your hardware for your own development work, but it's clearly not enough given that showstopper driver issues are going unfixed for half a year.

I'm not familiar with the particulars of OneAPI, but it's just a matter of rewriting CUDA kernels into OneAPI. This is pretty trivial for the vast majority of small (<5 LoC) kernels. Unlike AMD, it looks like they're serious about dogfooding their own chips, and they have a much better reputation for their driver quality.

Gemini and Gemma were trained on Google's TPUs.

He loudly gave up on AMD after they did not fix a blocker he had for 5+ months and gave him the runaround the entire time when he asked for the code to fix it himself. He is still shipping the AMD tinybox with huge warning labels.