I'm not American, but I know there already is an asteroid defence programme. I was listening to people involved in it on a podcast, Science Friday, the other week. Interesting stuff.

On the 2.5Bn years bit - yes, but don't start don't finish. We have no idea how long it will take to develop the techniques required for this, and there will always be a good reason to put it off until tomorrow. Needs to start and just become so embedded over time that future generations don't even question why we're doing it - it's just patently obvious to them. Won't happen in my lifetime, but then a lot of things won't happen in my lifetime that are getting worked on today and that's fine.

There's often a "we should fix problems on earth before looking to space" theme. We can be as equal, progressive and fair as we like and an asteroid still wouldn't give a damn and wipe us out anyway. Should we survive asteroid attacks, the sun will expand and burn the planet dry anyway.

We have to do both. Defence of the Earth (dramatic phrase, but see subject above...) has to be studied, funded and run. Along side that a long, probably multigenerationally long, programme of "how do we survive when the Earth is uninhabitable" including the ability to leave Earth and live elsewhere. These programmes are fundamental to long term survival of the human race.

I think one of the problems is that it all sounds very dramatic, big and sci-fi. But it isn't - we have direct evidence that the risk already materialised once and wiped out most life. We also have evidence of the expansion of the sun. All these things are certain, so we have to look at them as reality and not fiction.

Interestingly, I would regard never learning how to write PHP over 21 years as one of my greatest successes.

So AI ingests the text books to learn about the subject so it can...create text books?

Some is from London at least, because I ended up behind one of the training cars a while ago.

I'm sceptical, but I'm not on the "never" path. It was going through one of the most annoying parts of my drive - for those that know the area, just outside of St Pancras heading up to Pentonville Road. At that point you have a lane split, comically bad driving, buses overlapping the lane, psychotic cyclists ignoring lanes and red lights and pedestrians on a tiny sliver in the middle of the three lanes continually trying to kill themselves in ever more novel and entirely unexpected ways.

If they tested it there and they still want to launch - huh. I'll be very interested to see it.

Every time I see a goofy new feature from Mozilla, I'm reminded how Firefox started as Netscape without the bloat. That was a great product. I wonder what happened to it.

Why not PDF - releasing archive documents in an editable format isn't the way forward either. Obviously in this case it's more of a statement than a technical choice, but...yep, archiveable read-only is the way forward for things like this. They shouldn't have been editable to start with.

International trade wars are one thing, but reverting to Microsoft Works format is going too far.

The original Tron was fantastic... if you were an '80s computer geek. If your experience with computers starts in the '90s or later, it's hard to relate.

Tron Legacy went all metaphysical. Metaphysics is where science fiction goes to die, and die it did. Tron Ares is a fitting sequel to Tron Legacy, which is to say: hot garbage with nifty special effects.

I'm reading a lot of "it's simple, just get a...". If you read the article, it says they're "associated with an early Mac computer". That almost certainly means these a GCR formatted disks, and need a drive that can do variable speed rotation.

It's not impossible obviously, but it's likely the best way to do this is with a vintage Mac itself. Which then implies hooking up a mass storage device of some kind to that Mac so that it can be transferred to something more modern. So not super rare and impossible, but definitely fiddly.

Not sure what Intel contributes that they could change to closed source without harming themselves. Closed source doesn't make it into the Linux kernel and with very few exceptions doesn't make it into the Linux distros. I can think of few Intel product lines where that wouldn't be destructive to their market share.

Wow, you can click the link to my web page and make a veiled threat to dox me. I guess my estimate of your intelligence was... right on target.

The diagnoses were merged because the evidence had begun to suggest that they were different severities of the same ailment. If the evidence has begun to suggest that we're dealing with fundamentally different ailments then the diagnoses should be split accordingly. If not then you're shuffling names for the sake of politics and it's not a good day in science when that happens.

I've always wondered how well Slashdot's moderation system actually works. Thanks for giving me the opportunity to find out.

Hmm - that site mixes operating systems with SSL usage on the same graph. But the other thing is - it's stats about public facing internet accessible sites. The majority of Red Hat clients are RHEL are internal or data centre, non-public. I'm struggling to get a link that works, but the 2025 estimate is around around 43% market share, and I'm honestly surprised it's that low.