I'd mod you up if I hadn't already posted.
I'd mod you up if I hadn't already posted.
Once again, I think we can turn to Betteridge's law of headlines: "Any headline that ends in a question mark can be answered by the word no."
Haven't used KDE since the V4 release myself, but I still tend to suspect that Betteridge probably applies here.
Pretty much what I was going for, yeah. Needlessly, pointlessly pedantic.
Is Debian itself "Debian-based"? I mean, technically (the best kind of correct), it is Debian, which is not quite the same as being "based on" Debian. There's a lot of things I'd describe as being "based on Shakespeare" (like the classic movie Forbidden Planet), but I would not describe Shakespeare's plays that way.
(This is Slashdot; we're supposed to pick minor nits, yes?)
p.s. I realize I've violated the unwritten rules of slashdot by actually reading the article and commenting on what it says, instead of leaping to snap judgment based on the headline alone. In my defense, I actually read the article yesterday, before it was posted to slashdot.
I like the "to announce" part. Like, if they haven't announced it, why are you reporting on it? Maybe there's a reason they haven't actually announced it yet! Perhaps the data is tentative and admits of another explanation, which, on further review, will prove to be true. Perhaps it's simply one guy's wild-ass guess based on incomplete data.
Contacted by AFP, ESO spokesman Richard Hook said he is aware of the report, but refused to confirm or deny it. "We are not making any comment," he said.
Maybe, just maybe, there's a reason he's not making any comment? Like, they want to avoid making false statements in public and embarrassing themselves? Quite unlike certain (most?) Internet "news" sites which are perfectly happy both to make false statements and to embarrass themselves? "Who cares? Just give us those clicks!"
Anyway, this is pretty cool if confirmed, but at this point, I'm treating it with all the seriousness it deserves, which is approximately zero.
I still have one gopher site bookmarked, although it's been dead for a while (and modern browsers no longer recognize "gopher://" as a valid URI prefix). I simply haven't got the heart to delete it. Of course, even at the time I added it, gopher was nearly dead as a protocol in general. I mostly only added it because I was astounded that I'd stumbled across a still-running site, and that my browser, at the time, could still talk to it.
This story was simply posted to gather evidence confirming the fact that slashdot is full of old pharts, yes?
What are you on about? Two months? Closer to two years! LO started in 2010. OpenOffice was, as far as anyone could tell, a dead project at that time. The LO folks spent a full year cleaning up the code before making their initial release in early 2011. It was half a year later that Oracle finally (after a year and a half!) announced they'd be relicensing the old OOo code and donating to Apache. What was LO supposed to do? Stop everything and wait to see if this promised code drop (of code they'd already spent a year cleaning up, and another six months improving and adding features to) actually happened, and start over from scratch? That would be dumb!
Now, for a while there, things might have gone in all sorts of different ways. I installed both, and was happy to go with whichever one turned out the best. But AOO completely failed to attract developers. They had some strong support from IBM for a while, but IBM seems to have abandoned them at this point. They're down to a tiny handful of developers, and they currently have a major security bug (CVE 2016-1513), and they can't even figure out how to get updates to their users, even though they have a patch!
I've got no skin in the game. I like Apache too. Have several friends who are members of the Foundation. But it's clear to me at this point that LO has won, and AOO is a dead project walking. AOO has about a dozen people who have contributed in the last year. LO has hundreds. LO has more changesets accepted per day than AOO has in months! LO has backing from several major companies, most notably RedHat and Collabora. AOO lost their only corporate sponsor, IBM, over a year ago.
You use what you want to. AOO is gone from my system, and I don't miss it at all. Being supported by Apache doesn't mean much if you haven't got any developers, and can't even figure out how to get a security fix out to your users!
Well in the stable 2.8.x series you only have 8 bit support, not 16/32 bit as far as I know.
The 2.6 series had 8 bits per channel (32 bits per pixel with the standard four-channel RGBA). The 2.8 series added 16 bits per channel internally by switching to GEGL, but didn't modify the UI to take full advantage. The 2.10 series (of which 2.9.4 is the latest dev pre-release) fixes that.
Nobody uses 32 bits per channel. Standard computer hardware only supports 8 bits per channel. The only reasons to even include 16 bits per channel are: it is provided by some cameras, and you don't want to lose precision, and it makes some transformations less lossy.
Yes, first-past-the-post is a bad bad system, because it inevitably leads to situations like this. The sensible response, however, is not to pretend we don't have first-past-the-post, because we do!
The sensible thing to do, from a game-theoretic viewpoint, is to vote strategically (because that's provably the best thing to do with FPTP), and, in the mean time, try to get other voting systems accepted. My city uses instant runoff (IRV) instead of FPTP. If more cities used IRV or Single Transferable Vote (STV) or similar, then we'd be able to talk about getting the state to do the same, and if enough states switch, then maybe we could get the country to switch. Then you could vote for your favorite candidate without worrying that you'll case the worser of two evils to triumph. In the meantime, however, vote using your brain, not your gut. You're supposedly a nerd, unless you're on the wrong site, so that shouldn't be a difficult concept.
(And again, "vote strategically" does not necessarily mean "vote for the lesser of two evils". That's simply one possible--albeit frequently useful--strategy of many.)
In essence, what he's saying is: "We don't want a back door. We just want a back door!" Weakened encryption *is* a back door! And more than even the usual back door, it's a back door available to everyone! (Although that's true to some extent of every back door.)
I'd have to check the latest polls. First-past-the-post demands strategic voting. My state is normally guaranteed for what I consider the lesser of two evils, so my usual strategy is to vote third party, to help send a message. This year, though, I'm seriously worried about the spoiler effect, and might end up voting for the major-party candidate that most closely aligns with me, if the risk of the split vote ending up throwing the win to the other side seems too high. I've certainly never seen the risk this high, in the several decades that I've been voting. Especially with all those Supreme Court positions at stake!
Note that this logic should apply no matter which of the major parties you feel closer to, since we have two widely hated candidates on offer this year.
Why would anyone expect a CEO—even of a tech company—to have any idea about computer security? That's like expecting a POTUS to have a deep knowledge of battlefield strategy, simply because they're Commander-in-Chief of the US Armed Forces.
Yeah? And imagine what it would be like if the crossing were run by a private company with financial incentive to discover contraband. We'd have 100% "bend and spread 'em!" I'll take good old-fashioned government incompetence over corporate goons with a financial interest in fingering my bunghole any day!
Yeah, C++11 was a major improvement; much of the extra "complexity" actually ended up making a lot of things much simpler. The new "auto", the new lambdas, and the new "smart" for loops, between them, allow writing simple, straightforward code to do things which would have required a tangled, opaque, and fragile mess of barely-readable code in older versions of C++.
Rvalue-references and std::move are a little trickier to understand, but fortunately, they can provide benefits even if you don't use them directly, since they greatly improve the performance of the standard library container classes and such. Even if you don't change your code!
Happiness is twin floppies.