yakult

Alternative hypothesis: most 'secret family recipes' are in fact original. However, the average SFR never get passed around, because it produces average-tasting stuff.

The recipe on food labels come from companies that know the product inside-out, have incentive to help you optimize the taste/effort tradeoff, and maybe have spent time and money on research. They tastes better, so they're the ones people remember.

tl;dr: people remember a disproportionately high number of plagiarized recipes because those are the good ones.

New Atlassian logo is better than the old logo. Old one always looked more like an unimportant decorative element than logo.

I totally agree about the hate on rich markup replacing plaintext, though. Entirely counterproductive and stinks of excess engineers syndrome.

They're probably going for native advertising.

Sure, we CAN contribute. But what's in it for us?

Are we going to get a browser that caters to our own needs? No, evidently power users are no longer the target demographic.

Are we going to make a browser that we can recommend to our nontechie friends? No, I don't trust them to navigate all the opt-outs and dark patterns around your telemetry. I don't even trust myself to never misclick.

Is contributing going to win us goodwill from our collegues? No, you've alienated them too.

Is this about ideology, then? Are we building a browser for a better world? ..it would be much more convincing if you guys didn't fire your CEO over political speech.

And that's where we're at right now. Maybe the 97% or whatever non-addon-using demographic in your telemetry data will make up the shortfall in contributions.

The telemetry probably shows that all the power users opted out of telemetry.

Quite a dangerous tool.

What happens if I copypaste some random text into the file and press save, by mistake? Are there mechanisms to limit the damage?

Also, when I remove a hash and save, the file is gone permanently, even if I paste it elsewhere, right?

...or you could just scrap the whole idea and not bother with it.

This is true for the user, too. If the only viable choices are 'verify claims at great cost and no gain every few months', or 'use some other privacy-respecting browser', I am going to recommend the second.

what percentage of FF users on the planet do you expect could read a paper on differential privacy and actually verify those points, while understanding all the ifs and gotchas, and be able to tell if any of the arguments are wrong? What percentage of that elite group would actually be willing to devote the time and energy, for free, for every one of the thousands of softwares they use?

So, what are the odds that we'll have an AI that writes at GoT quality before George finishes the last book?

Look at it from a security-conscious user's perspective: I would have to verify that:

1. The concept is sound. 2. It is implemented as described. 3. It is implemented with no bugs. 4. Mozilla is trustworthy 5. Any third-parties Mozilla involves in this process are also trustworthy. 6. All of the above will remain true.

Doing this would take a tremendous amount of both time and expertise, if even possible. If every piece of software I use makes me do this every year or so, I would get nothing else done.

In practical terms, your argument is no better than just saying, 'trust us, we're good for it', regardless of the merits of your tech. And we know Mozilla baked Google Analytics into FF's addon page, so trust is in short supply.

1. Any data collection at all deanonymizes the user, cf panopticlick.

2. Frankly even opt-out is not acceptable. I can't recommend any software that peridically asks users for data access, since there exist non-technical users who have a nonzero chance of clicking yes to everything. If they are related to me in some way this compromises my privacy also.

Look at it this way: whenever you are running a program you didn't write yourself, you're running a bunch of commands you never checked. This is no different to, say, downloading a precompiled executable and running it, with all the same problems and tradeoffs.

>what nefarious things can one do knowing that some anonymous user happens to e.g. refresh pages a lot (or even, how can that data be sold)?

Page refresh pattern data would go a long way in deanonymizing the user panopticlick style.

Generalized version:

"Only <low status> people get <common condition that I do not have, or would like to signal I do not have>".

Once you see the pattern you'll see it everywhere.

You need access to expertise, equipment, grounds, peer competition. These all require money and/or connections.

Going pro depends significantly on the economics of supply and demand, if it's defined as 'finding enough people to pay you to do this for a living'. It's easy to go pro as a plumber. Not so easy when you have to compete vs 300000 people for a PGA slot.

I think the reason is that if he had used 'she', there would have been a shitstorm.

If you do a word-replace of all instances of 'he' with 'she' and 'his' with 'her', it becomes an entirely different article.

Nonesense, obviously the money is in selling recorded conversations to the highest bidder. Identifying spending patterns via image recognition of pets and babies would be up there also.

What's the current state of art in this direction? Is there a way to encode equations explicitly prior to training?

Academic code typically just has to work once or a handful of times, for a small number of highly expert users, frequently just for the author. Ease of update is of the essence - you'll rewrite most of it many times, as your understanding the problem change. You can use all sorts of ugly hacks so long as you get what you're after.

If any of it ever becomes commercially released or whatever, there'll need to be a complete rewrite that makes it usable and maintainable by people other than yourself. But most of the code will never get to that point because most of what you've done up until about a week ago is wrong and worthless, and the current, correct-until-next-week iteration is stuck together with duct tape.

Speed only matters on the infrequent hot paths, which is why Python is popular. The rule of thumb is nobody cares about speed / resource consumption until it needs to run on a cluster, but then you care a lot because cluster time is metered and simulations can get huge. Fortran is still fairly popular because many math libraries are on it and porting would require huge effort from a very small group of very busy people.

Most of the coders are not software engineers and don't know / don't follow best practices; on the other hand the popular best practices are not designed for their use-case and frequently don't fit. Versioning (of the I-don't-know-which-of-the-fifty-copies-on-my-laptop-is-the-right-one type) is a big issue. Data loss happens. Git/Github/etc has steep learning curve, but so does all the various workflow systems designed for research use.

This is a story about misaligned incentives and social status.

Having the special programs lowers the social status of the demographic it targets, because it suggests those people are not good enough to climb the ladder through meritocratic means. At the time they do lower the hurdle for the demographic seeking to enter the industry and/or get ahead, fairly or otherwise. In otherwords, there is a tradeoff between social status and initial growth.

Whether this is a favorable trade depends largely on where you sit. If you're high up already, you would likely prefer the programs to go away since they hurt your status by associating you with the low-status newbies while not giving you much. On the other hand if you've just submitted your resume on BigTechCo's website like a thousand others, it's much more likely to get read by a human if they have some kind of affirmative action program (at the cost of somebody else's resume not getting read by a human, of course). Once you get in, extra programs and mentorship would also be valuble.

Finally, from the company's perspective, it just wants the statistics-illiterate social media diversity mob off its back while deviating as little from its current HR formulas as possible (since they obviously worked well enough to get them big enough to have to worry about this). This means it'll optimize for visibility and photo-ops and photogenic suit-clad female interns in its annual reports.

Depends on whether I get a price better than my assessment of the asset's present value.

This is purely speculation, but judging by how gun-ho they were about OPEC production cuts, they may have much less oil than widely believed. It could be they would have needed to cut production anyway since the wells are aging, and they're just making the best of it.

There's this, there's the post the other day about encouraging 'openness' with MDMA, there was the article proposing 'fairer' jerrymandering reform that just so happens to shift all the votes in a single direction, plus all the hagiographies of our tireless selfless social media censors, I'm beginning to think there's a burgeoning industry working on novel side-channel attack against the electoral system.

The lesson there is that people have unrealistic budget and schedule expectations of large IT projects. Same with defense, medicare, etc.

Meanwhile in America: https://en.m.wikipedia.org/wiki/Solar_Roadways

Unlike that debacle, this might even make economic sense once you factor in the government rebates/sweetheart deals and the PR bonus. Alas, nobody gives me anything on the side for buying a powerwall.

Numbers don't make sense yet. Plus it takes up space, is a fire hazard, and I am doubtful the 10-yr+ warranties will actually hold up past year 5, especially for third parties.

Right now it's mostly for conspicuous consumption and virtue signalling. Those of us number nerds are waiting for the tech to improve and the price to drop more. A move off toxic/flammable lithium tech would be good too.

For large companies, I can imagine using new branch offices for a/b testing.

A while ago I read a comment here about a university that, instead of paving roads to a new campus, just had grass everywhere. A semester later they paved over all the parts where the grass died from all the foot traffic.

I wonder if something like that would be workable? Have easily reconfigurable panels and furniture, encourage the staff to arrange it however they like, wait six months, then standardize/remove pain points/make permanent?

I suspect that's actually part of the problem: a massive buildout of renewables without full accounting or planning for the complications to the grid, i.e. all the shock loads and lack of storage. Then the consumer is left footing the bill.

At the same time, we have signed long-term LNG contracts while simultaneously large gas projects are delayed by various red tape and doesn't look like it'll clear up any time soon. I read at some point Australia is importing LNG just to reexport it to meet contract requirements.

>It's easy to walk into someone's office and have a discussion.

Certainly less easy than joining in on a conversation you overhear while sitting at your desk.

What we have here is disgreement over the optimal 'activation energy' for social interactions that produces the best work. This parameter depends on all kinds of factors, e.g. the type of work and work culture. What we need is a workflow that optimizes this parameter for a given office.

IOW, somebody needs to come up with a cost function and run gradient descent on it.