No it hasn't, at least not under Windows. Just uninstall 0.9RC, delete the "%USERPROFILE%\Application Data\Mozilla" directory and install 0.9. It'll ask you if you want to import the Phoenix (0.8) profile.
However, I simply think that several crucial mistakes were made in 0.9.
Well at least the profile "trashing" was your crucial mistake, not FireFox's. Tip: never install "release candidate" software without backing up everything you're not prepared to lose. And always read the release notes.
Doc Searls (one of the bloggers affected) provides a bit more info.
Um, Doc Searls is quite clearly not one of the bloggers affected - his blog is still up. Clearly he's got special treatment, but how does that help the 2999 other bloggers who have no chance of seeing their data for another two weeks?
Unless you're proposing that not only should copyright for the title be dropped - its source code should also be made available so that it can be adjusted to work on current hardware and future hardware to come.
Not necessary. Emulators can take care of this in the same way that they are extensively used for old arcade games (which are another example).
Carmack did seriously consider using Java for Quake 3 (though not for the rendering engine, presumably). In the end he decide to include a C interpreter instead, the reason being portability issues with Java.
The author seems pretty stuck on extremely stretched "real-life metaphors". I never ever actually thought of files & folders as drawers in a cabinet, or webpages as pages in a book -- any artificial attempts to link these two quite separate activities are doomed to failure.
Exactly right. Metaphors break down, and tend to get in the way when they do. In UIs metaphors should be used to reduce the steepness of the learning curve, but should be abandoned as soon as practical and not pushed beyond their natural applicability.
In this case, the "drawers / cabinet" metaphor doesn't even match particularly well - it doesn't explain links at all, it doesn't map well to the deep hierarchies that are common in filesystems (what's that supposed to be - a drawer inside a drawer inside a cabinet?), and it doesn't explain removable media well. I've used computers long enough to want to think of my harddrive's contents as what they are - files, directories, and links. I want a window to be a view (that I can change) into the filesystem, not a representation of a specific directory. Any interface that gets in the way of that is a bad one.
The problem with this spatial mode Nautilus has is that it doesn't account for what people want to do. In probably 90% of cases a user opens a new directory they are finished with the old one and leaving it lying around is not the correct thing to do. As Jobs said, it makes the user into a janitor, constantly having to clean up unwanted windows. The author responds to this point by saying "you can use double middle click instead", but why have such an obscure operation for the common case, and why open a new window and close the existing window when replacing the contents of the current window makes more sense?
You're assuming 'limit' is less than or equal to 256 without checking. You're assuming that x, y, and z are set to nulls. You're not handling regional text (i.e. accented alphabetics) or unicode. You also haven't accounted for the fact that the Perl code returns the three bracketed sections of the match in variables.
I'm interested in when we can do this in a game in real-time. l(2hrs*3600secs/hr*60fps)/l(2) * 1.5yrs = 28 years before we see this in real-time (though that's using Pov-Ray, which could probably be sped up a lot if it's made into a game engine rather than a general purpose graphics architecture.
Certainly I agree. I'm just pointing out that some of the comments that have been made here imply that he was the only memeber of Bletchley that mattered and that is certainly not true.
And you're basing that on what? Turing wasn't the only person working on Enigma. Does the name Marian Rejewski mean anything to you? He was the Pole who figured out how Enigma worked and how to crack it. He did as much for the Allies as Turing. In fact Turing's job was to find a second method of attacking Enigma incase the Germans changed the procedures that allowed Rejewski's method to work. There is no evidence to suggest that only Turing was capable of figuring out that method. Other people had performed similar feats.
Even with Turing's method not every message was broken (or even intercepted). When the Germans changed procedures and even the design of Engima (in the case of the 4-rotor Naval version) the Allies often lost the ability to break the codes for weeks or months at a time. Often it was captured codebooks that allowed the codes to be read. Without Turing's work other ways to gain the required intelligence would have been found.
Even if the Allies had of lost the ability to read Enigma-coded messages entirely it is not clear that it would have lost them the war. It's extremely difficult to assess these sorts of scenarios, of course, but don't forget that Enigma intelligence was only one small part of the intelligence available to the Allies.
No Enigma cracking, we lose the Battle of the Atlantic.
Most Enigma cracking during the Battle of the Atlantic was based on captured codebooks, up until the start of February 1942. That is when the German navy switched to the 4-rotor Engima. Little progress was made against that until the capture of the new codebooks from U-559 at the end of October. Bletchley wasn't regularly cracking Enigma again until mid-December. So for 10.5 months during the most intense period of the Battle of the Atlantic no Enigma intelligence was available. Cracking Enigma was a big factor in winning the Battle of the Atlantic but it was not the only factor (radar was another for example), and it is not clear that we would have lost the battle without Engima.
I think what they ment was without him, Hitler would be drinking tea at No.10, but he did have a pretty big impact.
I know it's romantic to make Turing out to be the saviour of Britain, single-handedly winning the war against the Nazis, but it's not really realistic. I don't want to take anything away from Turing, who was a truly great man, but deifying him the way some around here are subtracts a lot from the achievements of the many other people who made significant contributions to the war effort. The fact is he wasn't the only genius at Bletchley, and if he hadn't been there they probably would have managed anyhow.
That is slightly over 5%. Sure, in many industries that would be very high but at a tech company - I am not so sure - and for a mature research organization that might be low (the drug industry or checmical companies).
I doubt it would be low. It would be low for the research divisions in drug and chemical companies but it would be higher than any of their overall percentages. Don't forget that the entire research division at those sorts of companies probably account for less than 10% of the workforce typically. Dow Chemical (for example) would almost certainly require PhDs for research, however they have over 46,000 employees. I doubt they have 9200 PhDs on staff. Pfizer, the largest drug company, have 122,000 employees of which 12,500 are in the research divisions. Not all of those 12,500 have PhDs though as that includes research management and many lab technician type roles (in fact browsing their job listings shows that few positions require PhDs, probably less than 10%).
I'm looking forward to Stirlings assault on solar, and wind power. After all the electricity itself can (and is) used by "dictatorial megalomaniacs in command of national economies, secret police and large armies" for a lot of nefarious purposes.
I'm not sure you've got a valid point here. The technology in solar, wind power, etc, are not directly applicable to nefarious purposes.
It seems particularly stupid in this case to say that we shouldn't use nuclear power because it might allow our government to attain the nuclear weapons. A valid argument against exporting nuclear technology perhaps, but not very convincing in denying it to ourselves.
But we can't "fix" global warming by merely switching to nuclear generation ourselves. A bunch of developing countries with either no or limited nuclear capability would have to be switched over as well. He does kind-of have a point here: to reduce the worlds greenhouse gas emissions by adopting nuclear power would mean giving it to a bunch of people we'd probably rather didn't have it.
Highlighting having the side-effect of copying is just unintuitive and often the wrong behavior. It's speculating that you will want to copy the highlighted text, but often times you want to delete the highlighted text without clobering your copy buffer. or maybe you might just want to highlight the text to mark your spot.
I agree, highlighting to copy is the UI guessing what the user wants to do, which is ok, but there are bad consequences if the guess is wrong, which is not ok.
Here's what I think Unix should do: ditch highlight to copy, ditch the distinction between primary and clipboard buffers. Keep middleclick to paste. That is a great use use of the middle button (under Win2K I loved that the Command Prompt would paste on rightclick). Keep Ctrl-C, Ctrl-V. Have standard rightclick context menu cut/copy/paste options as well. Basically Windows behaviour plus middleclick to paste. At the very least I should be able to choose how the system behaves.
TCP utilizes low-value waste by-products such as tires, plastics, sludge, municipal solid waste, paper, animal, and agricultural waste as feedstocks.
Note "low-value". They are not using stuff which has significant value as fertilizer. Also note that fertilizers are generally nitrogen based, and oil is a hydrocarbon. From the FAQ page: "What other materials are you producing?
We also produce gas, minerals and fertilizer". So you see the parts of the waste that are valuable as fertilizer are still being sold as such.
Colossus would however easily crack Enigma (sic) codes quicker than your over-clocked P4.
Tony Sale (the man who led the rebuild effort) wrote a simulation of the Colossus machine in Javascript. It is available here.
A description of the simulation is available here (PDF).
In Appendix 1, Sale notes: "as a result of this decision (to use Javascript) the programs are BIG and the simulation runs slower than the original Colossus. (about one second with a 600Mhz PC to scan and process 2,000 input characters, original Colossus, 5,000 characters per second)". Clearly an implementation written in a decent language running on modern hardware would be several times (possibly as many as 10 times) faster than the Javascript simulation and therefore faster than the original Colossus.
Funny thing about slashdot - people seem to think they know all about hardware because they know the difference between a MHz and a GHz.
Funny thing about Slashdot - people seem to think their opinion is fact without bothering to do any research.
Slashdot Mirror
For an example of this, just look at this thread.
In this case, the "drawers / cabinet" metaphor doesn't even match particularly well - it doesn't explain links at all, it doesn't map well to the deep hierarchies that are common in filesystems (what's that supposed to be - a drawer inside a drawer inside a cabinet?), and it doesn't explain removable media well. I've used computers long enough to want to think of my harddrive's contents as what they are - files, directories, and links. I want a window to be a view (that I can change) into the filesystem, not a representation of a specific directory. Any interface that gets in the way of that is a bad one.
The problem with this spatial mode Nautilus has is that it doesn't account for what people want to do. In probably 90% of cases a user opens a new directory they are finished with the old one and leaving it lying around is not the correct thing to do. As Jobs said, it makes the user into a janitor, constantly having to clean up unwanted windows. The author responds to this point by saying "you can use double middle click instead", but why have such an obscure operation for the common case, and why open a new window and close the existing window when replacing the contents of the current window makes more sense?
You're assuming 'limit' is less than or equal to 256 without checking. You're assuming that x, y, and z are set to nulls. You're not handling regional text (i.e. accented alphabetics) or unicode. You also haven't accounted for the fact that the Perl code returns the three bracketed sections of the match in variables.
X.org is now part of the FreeDesktop.org project. Keith Packard is very much involved with the X.org implementation.
Certainly I agree. I'm just pointing out that some of the comments that have been made here imply that he was the only memeber of Bletchley that mattered and that is certainly not true.
Even with Turing's method not every message was broken (or even intercepted). When the Germans changed procedures and even the design of Engima (in the case of the 4-rotor Naval version) the Allies often lost the ability to break the codes for weeks or months at a time. Often it was captured codebooks that allowed the codes to be read. Without Turing's work other ways to gain the required intelligence would have been found.
Even if the Allies had of lost the ability to read Enigma-coded messages entirely it is not clear that it would have lost them the war. It's extremely difficult to assess these sorts of scenarios, of course, but don't forget that Enigma intelligence was only one small part of the intelligence available to the Allies.
Most Enigma cracking during the Battle of the Atlantic was based on captured codebooks, up until the start of February 1942. That is when the German navy switched to the 4-rotor Engima. Little progress was made against that until the capture of the new codebooks from U-559 at the end of October. Bletchley wasn't regularly cracking Enigma again until mid-December. So for 10.5 months during the most intense period of the Battle of the Atlantic no Enigma intelligence was available. Cracking Enigma was a big factor in winning the Battle of the Atlantic but it was not the only factor (radar was another for example), and it is not clear that we would have lost the battle without Engima.And consider the source: (the scientist making the claims). He's got some pretty out there ideas.
Simply test all messages against a regular dictionary and flag anything with too high a percentage of misspellings as probable spam.
Here's what I think Unix should do: ditch highlight to copy, ditch the distinction between primary and clipboard buffers. Keep middleclick to paste. That is a great use use of the middle button (under Win2K I loved that the Command Prompt would paste on rightclick). Keep Ctrl-C, Ctrl-V. Have standard rightclick context menu cut/copy/paste options as well. Basically Windows behaviour plus middleclick to paste. At the very least I should be able to choose how the system behaves.
Which would still only be the first of the two flights required. They can't claim the prize on July 4.
What's the distinction?
In Appendix 1, Sale notes: "as a result of this decision (to use Javascript) the programs are BIG and the simulation runs slower than the original Colossus. (about one second with a 600Mhz PC to scan and process 2,000 input characters, original Colossus, 5,000 characters per second)". Clearly an implementation written in a decent language running on modern hardware would be several times (possibly as many as 10 times) faster than the Javascript simulation and therefore faster than the original Colossus.
Funny thing about Slashdot - people seem to think their opinion is fact without bothering to do any research.