The entire process is broken. Companies can't find qualified people, so they turn to a swarm of useless recruiters who just want to get a warm body in a seat. Job postings never tell you what the company actually does and just describe a generic set of responsibilities, most of which don't actually even pertain to the position in question. Prospective employees have no way to convey their true capabilities and must rely on doing well in an arbitrarily complex interview process that is no better than flipping a coin in terms of the quality of candidate it hires. The questions asked in the interview never represent the state of the team's code or culture, either. Maybe they had a coding question for you to design and implement a parser in half an hour, but I guarantee you their code base does not even use data structures unless they're passing all their data around in hashes of hashes. And everyone accepts this, because what else are you going to do?

You'd think one company could come along and do everything just slightly better and blow all their competition out of the water, but so far this hasn't happened.

There is no silver bullet. Agile can't fix your broke-ass corporate culture, open source won't magically make fairies and unicorns fly out your butt. They're just keywords seized on by inept managers trying to mask their own deficiencies.

Today's Captcha: Click on the tiles that have your social security number in them!

That's how I always wanted to die! Well, that or supernova. That'd be awesome. Briefly.

The satellites are where they need to be, though. In geosynch, they're servicing one specific area of the planet. You couldn't consolidate all the services of all the satellites in the various geosynch orbits into one location, as that location wouldn't be able to service more than a tiny fraction of the planet. Same thing goes for lower orbits -- Musk's internet satellites need to move around the planet so that each one can service a tiny little region. If you move them higher, service latency increases. And you have to move relative to the earth that low. So in both cases, you need lots of satellites in individual locations.

If you wanted to offer a bunch of different services in one location, something like that might work. But you'd still want to build a custom satellite rather than a modular system, because getting mass into orbit is very expensive, so no one's going to want to launch any more of it than absolutely necessary. The infrastructure for a modular system would be an unnecessary cost to launch hardware that might never actually get used.

Something like that might work for the geosynch satellites out in parking orbits after they reach their end of life, if it ever starts getting too cluttered out in that space. But people don't tend to care about what happens to the hardware in parking orbits. The story's Northrop project aims to extend the life of some of that hardware, but it still has to be in reasonable working order. Most of the satellites that would need something like that would by that time have been in space for a decade or so and are expected to start malfunctioning due to the radiation. So while this is a neat accomplishment, I expect it to be somewhat rare.

It could make sense to have a satellite manufacturing lab out there somewhere, though. If you can build them in space, it'd cost a lot less to move them to the correct orbit afterward. But you'd have to make your own fuel at that location, otherwise the cost of launching fuel to the lab (to fuel the satellites they would be building there) would be greater than the cost of launching all those satellites individually.

Space stuff gets done the way it gets done because it's the least expensive way to accomplish those goals. When you're talking hundreds of millions of dollars per launch, the least expensive option is always going to get chosen.

I feel like the S3 was the pinnacle of Samsung engineering. After that their phones just got clunky, awkward to use. They are terrible both at being phones and at being the general purpose computing device I dreamed of having in my pocket a couple decades ago. Despite having hardware orders of magnitude faster than my desktop back then, it manages to be in every way a much less useful device. I'm thinking of just building my own phone with a raspberry pi that that just uses asterisk to connect to a VOIP provider over wifi, since all the current consumer phones seem to be awful.

Nothing happened again in France today. Life sure has been peaceful since Goggle had to start paying news providers!

'e's not dead! 'e's pinin' for the fjords!

If someone invented an AI that you could just talk to and describe what you want, most of the companies I've worked for wouldn't be able to make it work for them. Most of the time I have to use my familiarity with their business model to figure out the requirements they didn't give me, in order to keep their software running. The requirements that you actually get, assuming you get any at all, look like a third grader wrote them.

I made a thing. My code is well commented, has unit tests and I wrote a ton of documentation for it. Anyone who's not a fucktard should have no problem maintaining it. Eyes the fucktard the company hired to maintain it after I left. You know who you are.

AI's gotten pretty good at some things, but AI is not good at tasks that require a lot of seemingly unrelated information. If humans are bad at a task because it's computationally complex, AIs are going to suck at that task as well. If humans are bad at a task because it's a lot of boring pattern matching over and over again, then AIs will likely be good at it. Looking for video frames in video is pretty easy these days. Accurately predicting movements in the stock market is not.

So far everything that I can find on the subject says that China is handling the pandemic far more competently than the US administration has with four times the population. It also appears that the administration and members of Congress have known that there was a deadly disease on the way in early January and nothing was done at that time other than some stock sales. So I'm guessing they're just looking for someone else to blame at the moment, so that people don't blame them for the hundreds of thousands of US deaths to come.

Well you have to consider the context! Sure, 100k is a lot. Sure, a lot of those are completely unnecessary and due to stupidity on the part of someone or other. But compared to the millions who are actually likely to die as part of all this, yeah, just 100k would be a pretty great job. Also consider that although the administration's response has been non-existant at best and dangerously misleading at worst, other people are being mind-bogglingly stupid as well, so you can't really blame all the president for all the stupidity going out there. That'd be like blaming the sun for all the light on earth.

Also, keep in mind that the USA elected someone who was immediately clearly unfit for office. Even if we didn't vote for him, personally, we obviously didn't do enough to keep him out of office. We all should have donated more and volunteered for other campaigns. So honestly, we have no one to blame but ourselves and most definitely not him. This should in no way cause us to reevaluate the way our system works or is run. The pandemic is no time for thoughtful introspection or realizing that most of the world is handling it better than we are.

1. I wasn't there during that time, so I'm not sure of the version of Windows. I'd guess WinCE, just based on the fine judgement displayed in the company's other operations.

2. Apparently not. They'd just send it up a bunch of rotations to make, and it'd just make them at the appointed times.

3. Yah, ground systems was on Linux. At least some of their early satellites ran Windows, I'm not sure what the more recent ones did.

4. Oh yes, the satellites apparently had an atomic clock, same as the GPS satellites, IIRC. However, if the time coming from ground system said to make three specific rotations at a specific time, they'd just do it, and ground systems was a ridiculously complex set of systems designed by multiple teams, with no data validation or error handling between them.

5. I'm not really an expert on satellite hardware myself, but my understanding is that many satellites don't even have an OS, just dedicated hardware with code written right on the iron. Probably assembly or C. Or woven in, if you look at some of the early Apollo stuff. Some of their computers looked like knitting projects. You might see VxWorks, some variant of Windows and Linux on a few of them. Back in the heyday of the various commercial UNIXes, I imagine someone could have tried to license a UNIX to use. IBM was big on PowerPC for a while and I wouldn't be terribly surprised if AIX made it to space on some system or other.

Oh, I wasn't trying to imply that company was competent. They gave us all company T-Shirts one time, then almost immediately sent out an Email saying never to wear their T-Shirts because it'd make us a kidnapping risk. They also used to say, proudly, I might add, that if their storage provider's prices were one penny higher they would not be able to afford to buy the storage and if it were one penny lower the storage provider would not be able to afford to sell to them. By now anyone who's ever been employed at that company will know who I'm talking about.

Want some more? They knew their entire process was crap and couldn't bring new products to market in a reasonable timeframe. They adopted Agile something like three times while I was working for them, but they never gave anyone any time to get used to the process. They just kept throwing new processes at the problem every other month, to the point where developers just started ignoring them because the process being pushed this month would just be scrapped and replaced again in a month. They finally settled on getting their entire development team together off-site (A couple hundred people) for a week every quarter so that teams could work out what they needed to do in that quarter and badger the other teams into committing to do the work needed to support that. Oh yeah, good memories there.

Anywhoo, on behalf of that company, I'd like to say, please don't kidnap their employees. Anyone with half a brain left when they went public and anyone you could kidnap now would just set your country's space program back by a couple of decades.