Well, it probably hasn't hurt valve yet. But it probably will impact it eventually.

It is not hard to see that there is a good chance the computing landscape will have more ARM based system in the future. Mac laptops are ARM based, and MS produced surface with ARM for a long time now. The nintendo switch and switch 2 are ARM systems.

It seems reasonable for Valve/SteamOS to see a shift to ARMin gaming as a long term threat/opportunity.

Obviously you need software, like a mouse driver.
But, shouldn't all these features essentially be buttons. Shouldn't the OS be providing standardized interface to this?

I left windows for linux such a long time ago that news like this confuse me...

My wife shares a subscription with her mother. There are tons of things we get at costco that are not perishable quickly and that are hard to get at that price anywhere else.

Cheese is really cheap but it is bulk. We split it off when it gets in.
Coffee is also quite cheap for the quality you get.
Paper products are also priced cheap. Once again it is bulk; but if you have space to store, it is pretty neat.

Not all, but some of the meat are cheap as well. We usually buy one meat bulk. Prep half and eat it over a couple days and freeze the other half.
Both my wife and I cook a lot. So an i take of bulk food isn't that bad.

yeah, no. The best conference were always international in tech. Maybe more often in north america, but they always have been rotating international locations.

I attended 3 conference this year. And the foreign attendance was half of what it usually was. Steering committees of conferences I have talked to are looking into non US venues for all but one (which is by designa US based conference.)

same here I still use old cards. They work fine for the little bit of rendering, gou computing, and ai inference I do.

... but it is evil when a private open source project does it?

I see!

I've been playing with these genAI system both as code producer but as helper on various tasks.
And overall, I find the models quite brittle unless they are fine tuned on the precise task that you want.

The main problem that I see is that the tool is fundamentally a string in string out. But the strings could be absolutely anything including completely insane things without proper fine tuning.

Today, I am writing a simple automatic typo correction tool. The difficult bits are making sure that the tool didn't crap out. I mean, it is easy to check you actually got an answer from the tool. The problem is that sometimes the tools will tell you: "Sure, I can fix typos. Here is your text corrected: ". Ans so you have to toss that output out probably. But how do you figure out that it shat the bed? Well, you can't really, it is just as hard as the original task in some cases. So you bake various heuristics, or you get a different LLM to check the work of the first one.

At the end of the day, you really can't trust anything these tools do. They are way to erratic and unpredictible. And you should treat any of these tools as being possibly adversarial. It's exhausting to use really.

There was an article in communication onf the ACM this year about how their model was sustainable.

Basically, their income come is going to shift to publication fees and conference registrations. There is little reason to think it is not sustainable.

I don't think that member subscription was a major source of their income.

I had a roomba 15 years ago; it was very good. it lasted me about 8 years. I've had a couple replacement since and they were not as good. They crapped out A LOT quicker.

Now they are also a third of the price, so I don't care too much. But an OG roomba at a decent price would be nice.

AI valuation for sure!

Pokemon cards have been a stupid market for a very long time. There is no reason to think it will crash anytime soon.

Labubus are still just getting started.

AI valuation if they bust will bust before the hardware refresh cycle. So it seems it is the only one with a clear clock coming.

Most universities I interact with organizes workshops to train instructors and has money set aside for professional development of instructors

Universities tend to not care much abiut junior faculty getting trained in teaching. Because they want them to write papers and grant. Not teach atudents really.
And once their research goes down a little bit, the university wants them to focus more on teaching. But by then you have trained the faculty not to care.

It is sad really how poor incentive structure in universities yield poor education for students some times.

I don't agree that time limits are meaningless. There are plenty of tasks you should be able to accomplish within a reasonable time limit.

Every semester I have students telling me they knew everything but could not complete it in the exam time. Yet A students were done within half of the exam time. The time limit is not meaningless.

Fundamentally, all tasks are bound in time. If I contract a painter to repaint my bedroom one color and they tell me it's gonna take 7 months, I'm getting a different crew!

Now that time limit should not be too tight in how you test students. But a time limit is not meaningless.

(I teach CS in college and grad school.)

Yeah, that's essentially what I do. My typical exam time is 75 minutes because of the way my university schedule is setup. My exams usually do not require more than about 35 minutes to do. You can get a good prediction for peoples grade based on when they leave.
If they leave before 30 minutes, they are likely getting F; leaving between 30 to 45 minutes is likely A; leaving within 40 to 55 is likely B; leaving before 70 minutes is likely C, staying to the end are usually Ds and Fs.

The time angle only make sense in orders of magnitude. I am teaching database this semester. I have a few SQL questions on my exam. It should take you about a minute or two to figure out the question and answer it. If you are particularly slow/not too practiced, it might take you 4-5. But it shouldn't take 10.
I often see students saying "I could figure it out, but it would have taken me much more time". Then really it means you are not practiced enough, so I still feel the grade is fair.

> professors "struggle to accommodate the many students with an official disability designation,"

I wonder how they struggle to accomodate? At my university, if you get extra time, you take the test in the testing center. It is completely free to me. Actually, I'd rather ALL my student take it in the testing center. Then it give me time to work on my next module or research paper.

(I teach CS in college and grad school.)

You can't practically give unlimited time. An unproctored test will see massive cheating on the test. So if you want the output of the test to somewhat resemble the skill of the student, you need a proctor. And that proctor needs to go home eventually.
By the constraint of my university, most of my test are 75 minutes long. The way my tests are built, you probably should not need more than 45 minutes to answer everything. Yet, I always have students staying the entire time that is available.
During exam week, we have to give a full 150 minutes. Usually A students leave within 30-50 minutes. B students tend to leave within 40-60 minutes. C students will stay about an hour and a half. And D and F student will stay the entire time staring at their exam without having written anything on it for the last hour and a half; but still they will stay another 30 minute.

So unlimited testing time is just adding expenses and inconvenience without benefiting students at all.