johansch

54 ms·

211.
by johansch 9y ago
Institutional incompetence or some stupid "Disney Vault"-like reason?
212.
by johansch 9y ago
It is a bit funny how Intel always jumps onto every single computing trend, and then never ever makes any money from these bets in the end. How long has this been going on? At least since the mid 90s.
213.
by johansch 9y ago
This NYT reporter filed the first article on the topic: https://twitter.com/daiwaka/ His report (and tweeting) was so partisan that I got really concerned with the integrity of the NYT as a whole. This new piece restor
214.
by johansch 9y ago
Wow. A balanced mainstream story on the topic.
215.
by johansch 9y ago
How can you tell?
216.
by johansch 9y ago
I'm just now holding a 1996 Intel Pentium 150 MHz CPU (A80502150) with a ceramic encasement and gold-plated pins. I think it will survive for a very long time. http://www.cpu-world.com/CPUs/Pentium/Intel-Penti
217.
by johansch 9y ago
(Warning: Really loud screaming. Do mute your sound before watching this.) https://www.youtube.com/watch?v=qydSKppGRD8&t=38
218.
by johansch 9y ago
This on top of the various disappointments (way too sparse interior (saying that Model 3 drivers won't need a dashboard because of the $5k "self-driving" feature... facepalm ), way too expensive and needlessly grouped upgrade
219.
by johansch 9y ago
As an example; it does a very neat impersonation of the Amstrad PC1512 (my first PC, back in 1989) including a faithful rendering of its unique 16 color CGA variation. And unlike Dosbox, which doesn't try to run things at the proper sp
220.
by johansch 9y ago
Related: I've been impressed by the quality of PCem recently: https://pcem-emulator.co.uk/ (Windows, Linux. Runs well on Mac via Wine.) It emulates lots of specific pieces of historic PC hardware. Not cycle-accurate, b
221.
by johansch 9y ago
Sort of related, a little more well known and also not very well advertised: PC Magazine (Feb 1982 - Jan 2009): https://books.google.com/books?id=V2588uIxmAQC Infoworld (Dec 1978 - Mar 2007) https://books.google.
222.
by johansch 9y ago
> or there's just one Chinese whitebox manufacturer Perhaps these guys: http://www.winnsen.com/
223.
by johansch 9y ago
I don't really think we can blame IBM on this one. From what I can tell, they followed the specs. The problem was that the authority overseeing lots of very sensitive data forgot to specify that the individuals working on this project
224.
by johansch 9y ago
All-right, I have to ask; what is the correlation you see between that movie and these events?
225.
by johansch 9y ago
Well, one person was already fired for buying this kind of IBM, but that was just the director general of the authority that leaked the data. The thing most swedes have an issue with is that the government tried to silence the leak for like
226.
by johansch 9y ago
I guess the most likely scenario is that the current PM tomorrow at 10am will announce that the current goverment has resigned and that it's then upon the Speaker of the Riksdag/Parliament to try to form a new government that will
227.
by johansch 9y ago
You are assuming I don't want to profit on them being evil.
228.
by johansch 9y ago
This is the point where I'm smiling happily because I moved from being an FB user to being an FB stock owner. I feel like this pattern could be applied more often in real life... some company is being "evil"... don't jus
229.
by johansch 9y ago
I wonder if something clever could be done to automatically switch behavior over time based on keystroke timings, suggestion selection behavior, etc. Maybe it would just end up feeling random and unreliable though.
230.
by johansch 9y ago
That must be even more excruciating to watch than my mom, who types the entire search string while looking at the keyboard. If I try to hint and say, "look, there's an autocompletion thing that has already guessed correctly what
231.
by johansch 9y ago
In my experience these slow typists also tend to look at the keyboard while they are typing - not at the screen. Thereby completely negating the benefit for this group of people. I guess that for the intersection of slow typists who look at
232.
by johansch 9y ago
Opera has supported this abnormal behavior for like two decades now. I remember being shocked by how many tabs the Opera core browsing/rendering engine developers used to have open on their desktops when I joined Opera back in 2004. I
233.
by johansch 9y ago
Gah.
234.
by johansch 9y ago
> The best numbers I could find was ~60% for Sweden A year later (October 2016, rather than October 2015): 66.4% https://www.pts.se/sv/Nyheter/Pressmeddelanden/2017/Sju-av-t...
235.
by johansch 9y ago
In the same theme: moving very large amounts of data between GC and AWS, in either direction. I wonder what it would take for them to make this cost zero. Legislation? It likely wouldn't be very expensive to get a very high-capacity fi
236.
by johansch 9y ago
I guess the 100x means that they are using LISP? Or maybe the other way around? Not sure. Meanwhile we are told that HN has excellent protection against upvoting rings. How the heck does this reach the front page in an organic way?
237.
by johansch 9y ago
> Luke Gorrie (Australia - Sweeden - Switzerland) Maybe first learn how to spell the names of the countries.
238.
by johansch 9y ago
I should have known, an organization as complex as 3GPP couldn't not already have invented this :)
239.
by johansch 9y ago
Maybe they were too busy micro-optimizing engagement rates using a very large amount of A/B experiments and got stuck at a local maximum?
240.
by johansch 9y ago
Implementations for something as "innovative" as this are bound to be spotty early on. Is there even pre-existing infrastructure/services/software support in typical networks for stunts like this or was this something th

More ›