Last time I checked in Russian shops, it was $3 per CD. I mean, whatever fits on a CD cost you $3 to purchase, be it a full Windows+Office distro or a [$6,000] collection of professional image- and video editing tools.
What is funny, this seems to be a kind of legally sold product there:), i.e. with warranty, receipts, and even somebody's copyright stamped on the CDs (not Microsoft's copyright for sure). There are no policemen chasing the stores, no nothing. Software is sold on every street corner.
As far as I know, Microsoft has abandoned all attempts to stop this (yes they tried). Maybe it's a business strategy that actually works for them? "Close the eyes at illegally sold software, they are hooked anyway."
These Xerox/ Tektronix color printers are OK, but not very reliable. My department has one (8200N) and has had its predecessor for the four previous years. When the predecessor was nearing the end of its four-year service, it started to break in multiple places, jam paper, print stripes, etc. The mechanical failures were due to the plastick inner construction that eventually got worn in places. We threw it away and bought a new model. It has seemingly the same inner construction, including the parts that broke in the old one, but now we've got a three-years warranty parts included. I don't know why the IT folks chose to buy the same animal again. Perhaps, there wasn't any better option for color printing.
Also, it's somewhat slow. It takes about 15 seconds to print one side of a sheet which is has already processed (i.e. 30 sec for a double-sided sheet), and much longer than that for most of the first sheets that have to be processed. Our configuration doesn't have a hard disk, so it can't "quick collate" multiple copies, either. When I'm saying that it's slow, however, I'm comparing it to our huge laser monochrome printer that churns out a double-sided A4 sheet in under two seconds. Now THAT's fast enough for most of my printing tasks.
Ditto about brownish color on grayscales at high quality. This is because it uses color ink for all gray tones in this mode, resulting in smoother appearance (more weaker dots) but overall brownish color for gray. In the same high-quality mode it lays more ink on color images resulting in a better look for them. I wish it could tell color elements on the page from grayscale elements and include an option to print pure grayscales with black ink, uncoupling the two features. Alas, it merges them in just one mode.
I don't know about the costs. It's free for the employees.
"Obscure British "preprint physics archives" may be better said as "resource that includes not peer-reviewed publications" (and not editor-reviewed, either).
Substituting qualitative comparison for a quantitative one doesn't work, because it ends up stretched to the extremes and degenerate into zero-information marketing speak. Even cameras with the same number of megapixels give substantially different image quality. The reasons are different color matrices, image compression, optics, readout, sharpening, etc.
About your picture size: you either define "acceptable quality" in terms of DPI, or don't get any meaningless comparison at all.
We pay around $1 per liter of liquid nitrogen produced by a large machine installed in a basement (basically a refrigerator on steroids + separation column, it produces LN from electricity and air). This is considered expensive and is only used when our LN line from a large outdoor tank is not running. The tank gets filled cheaply from a huge cistern truck (the kind that crashed and temporarily froze Terminator 2 in the movie); the truck comes every couple weeks or so and refills the cistern.
A 25 liter dewar ($25) is already too heavy to be comfortably carried by hand.
I can imagine smaller LN machines should exist, down to the desktop size. These are probably too expensive for a household gadget, though
Well, maybe. I might not understand what things the nested loops we wired in LabVIEW actually included. It was not me who did it, but our near-full-time LabVIEW programmer. She spend some time trying to optimise the loops, and concluded it couldn't be done any faster in G.
Even though I've never programmed in C before, the first code with the same nested loops that I managed to get to work, worked fast.
The thing we needed to do was trivial, like accessing a port in a loop, or maybe sorting through a large integer array with mostly zero values, picking non-zero values and stuffing them into another smaller array (it was raw detector data in quantum cryptosystem, if you know).
LabVIEW (or, more precisely, the G language) is extremely slow comparing to less fancy, so to say, programming languages. Any routine data processing that you have to program step-by-step and that has to deal with a fair amount of data (e.g. a large array), will take forever.
I've had to embed several C routines in such places (through the Code Interface VI), and has sometimes been wondering what's the whole point with LabVIEW then...
Another observation (by a colleague of mine) is that he's never seen a program in LabVIEW that looks like a finished application. You'd always have to check first the initial paremeters scattered over different windows and in the source, and then start and stop the program with the Run and Stop buttons on the tool panel. The G language certainly encourages quick-and-dirty programming, even though it is in theory possible to create a standard stand-alone application in it.
I agree. For most sites, it's true, or should I say it's up to them to decide what browsers to support.
There is, however, a class of sites that are mission-critical. Such an important thing as controlling your domain name or finding information on the site of a TLD authority, must work in any browser, down to those used by 0.01% visitors.
Netscape 4.x is still used by perhaps one or two percent visitors, while requiring Flash locks out a hell lot of them.
Besides, not working in Netscape 4, while often being nearly meaningless in itself, is actually a reliable sign of further troubles with your Web design. It's just a correlation that for some reasons exist in today's Web design.
Any TLD authority whose site doesn't work in Netscape 4 looks cheesy to me... (reach for IE) and the one that requires Flash definitely won't be on my list. Period.
Don't be mistaken, these spam conferences are closely watched by the spammers as well.
Two weeks ago, I read A Plan for Spam article from the last conference, announced on Slashdot. There, the author describes spam-of-the-future as "some completely neutral text followed by a url".
Voila, the future has come. Yesterday I got a short message in Russian, in friendly tone, with an URL. Just like the ones I sometimes get. I'm a webmaster of a site with diverse content, and strangers sometimes send me stuff like this for news etc. There is absolutely no way to tell whether it's a spam or not without visiting the URL.
While the developers wrestle with one strategy and openly discuss the remedies, the spammer sees it and picks the next strategy, always ahead of you! Who benefits more from these conferences, good folks or the spammers?
One fix I'd propose is to stop publishing and webcasting the conference stuff. Then the spammers would have to attend in person. You know what happens next. A spammer surrounded with angry geeks:)
I'm not a space expert, but the most sensible thing for my country (I'm Russian) would be to fully support ISS operation with its Progress and Soyuz spacecrafts, until the things are sorted out with the Shuttles. Perhaps cough up some extra cash on the Russian side, yes. That would also be a politically correct thing to do.
This would mean the construction activity is halted (Shuttles were to deliver most/all new modules), but at least the station can be operated in its current configurations for the time being.
I view the dual delivery systems (STS + Russian crafts) as a partial redundancy built into the ISS program. Don't we now have the exact case when this redundancy should be used?
Good you mentioned pair.com. The way they deal with sudden surges: the day of highest bandwidth usage is discarded before calculating month's average and billing the customer.
This way, it automatically keeps everybody happy until the customer starts overusing the service in a systematic way, which is another story.
A lot of people complained about Flash, or lack thereof in the list of mistakes. The reason is simple: Nielsen has been hired by Macromedia to work on Flash usability.
Flash is apparently a problem on the Web. Either he excluded it from the list intentionally, or he believes the use of Flash is going to improve radically soon enough, now that he works on it:). Either way, he is biased and doesn't disclose it.
Well good. My browsing improved a lot since I deleted the Flash plug-in. I still don't miss it.
About half of this collection of Russian anthems originated from now-defunct Napster. I believe that the collection is now one of the largest, and linked to by many researchers.
If my university prohibited Napster, as some othes Scandinavian schools did, the collection would probably have never started.
Worse than that, I would never know first-hand what P2P is. This is about academic freedom: you should be allowed to test whatever darn new thing is out there, for whatever reason, otherwise the school lags behind. What you use it for, is your responsibility, of course.
Oh yes, I'm first-hand aware of the associated headaches (cleaning up the lab computers from those pesky money-generating add-ons that pop up an ad at the timing-critical phase of your data acquisition:-).
Unfortunately, the paper is in a somewhat particulate state: you have to download the text in MS Word format and the image separately.
Also, by poking around the (unprotected) Kurtsiefer's home directory, you can have a lot of images of the experiment, experimenters, and the beautiful surroundings.
My congratulations to Dr. Kurtsiefer for finally getting this difficult experiment to work. He's been trying to finish it for about a couple years, if I recall correctly.
Sometimes people do something similar using white text or text the same color as the background.
This is probably detected as spamming by the search engines. White and/or tiny text has been the pr0n way to boost ranking. You can't count on every search engine to correctly recognize it and index the white text with low priority, instead of ignoring it or, worse yet, firing your entire site from the index.
Too bad search engines are dropping them altogether. Keywords tag used to be an appropriate place to put misspellings, alternative and supplementary terms that should not appear as a visible text on the page.
Without keywords tag, you are left with e.g. this solution (scroll down to the bottom of the page). Not pretty, but search-engine compliant, huh?
Perhaps a better way would be to index these tags with low priority, as some search engines still do. This way, the keywords would only matter if there aren't many other pages with them (misspellings and rare terms), or in conjunction with visible text (variants and attributes). Well, a search engine can check misspelling of common words, but not rare terms and proper names. Both ways, the tags would be hard to abuse while useful in certain searches.
The laziness is working against this (why bother with something which is not visible on the page?), but without meta tags the Web is becoming dummier, in a way. Hope the search engines will master technology to replace them, but it's not quite there yet!
For those who can't read Norwegian, here are my favorite two bits from the story:
Nigerians ask a Norwegian to reimburse their plane tickets ($5500), which look obviously faked to a German man present at the conversation. The man goes down to the hotel reception at Bremen, calls the police and says "There is a Nigerian fraud going on". The police replies "We've already given so many public warnings on this. We do no babysitting until REAL CRIME happens, man".
Two months later, the Norwegian firm and a Norwegian bank (the latter unwittingly) have already been out half a million dollars, and are sending more. The police is alerted and arrests everybody.
FTP speed testers, beware of TCP/IP limitations
on
How to Test Your T1?
·
· Score: 5, Informative
The cap on FTP speed with high-bandwidth lines is usually imposed by round-trip time (i.e. ping time) and window size (a setting inside your TCP/IP client/server), NOT by the line performance.
Those who test by FTPing large files and watching the transfer rate, should understand these limitations (kindly explained to me by J.Spencer Love).
I had a similar problem trying to host a large-bandwidth video clip. It turned out the bandwidth of my 10Mbps line did not saturate at all (in fact, it was utilized at mere 5%), so neither did the trans-Atlantic connection. The bottleneck was the internal buffer in client and server software.
This also means you may not need that much bandwidth to push the speed of your FTP/TCP-based tasks to its limit.
Slashdot Mirror
What is funny, this seems to be a kind of legally sold product there :), i.e. with warranty, receipts, and even somebody's copyright stamped on the CDs (not Microsoft's copyright for sure). There are no policemen chasing the stores, no nothing. Software is sold on every street corner.
As far as I know, Microsoft has abandoned all attempts to stop this (yes they tried). Maybe it's a business strategy that actually works for them? "Close the eyes at illegally sold software, they are hooked anyway."
Also, it's somewhat slow. It takes about 15 seconds to print one side of a sheet which is has already processed (i.e. 30 sec for a double-sided sheet), and much longer than that for most of the first sheets that have to be processed. Our configuration doesn't have a hard disk, so it can't "quick collate" multiple copies, either. When I'm saying that it's slow, however, I'm comparing it to our huge laser monochrome printer that churns out a double-sided A4 sheet in under two seconds. Now THAT's fast enough for most of my printing tasks.
Ditto about brownish color on grayscales at high quality. This is because it uses color ink for all gray tones in this mode, resulting in smoother appearance (more weaker dots) but overall brownish color for gray. In the same high-quality mode it lays more ink on color images resulting in a better look for them. I wish it could tell color elements on the page from grayscale elements and include an option to print pure grayscales with black ink, uncoupling the two features. Alas, it merges them in just one mode.
I don't know about the costs. It's free for the employees.
"Obscure British "preprint physics archives" may be better said as "resource that includes not peer-reviewed publications" (and not editor-reviewed, either).
Substituting qualitative comparison for a quantitative one doesn't work, because it ends up stretched to the extremes and degenerate into zero-information marketing speak. Even cameras with the same number of megapixels give substantially different image quality. The reasons are different color matrices, image compression, optics, readout, sharpening, etc.
About your picture size: you either define "acceptable quality" in terms of DPI, or don't get any meaningless comparison at all.
We pay around $1 per liter of liquid nitrogen produced by a large machine installed in a basement (basically a refrigerator on steroids + separation column, it produces LN from electricity and air). This is considered expensive and is only used when our LN line from a large outdoor tank is not running. The tank gets filled cheaply from a huge cistern truck (the kind that crashed and temporarily froze Terminator 2 in the movie); the truck comes every couple weeks or so and refills the cistern. A 25 liter dewar ($25) is already too heavy to be comfortably carried by hand. I can imagine smaller LN machines should exist, down to the desktop size. These are probably too expensive for a household gadget, though
Even though I've never programmed in C before, the first code with the same nested loops that I managed to get to work, worked fast.
The thing we needed to do was trivial, like accessing a port in a loop, or maybe sorting through a large integer array with mostly zero values, picking non-zero values and stuffing them into another smaller array (it was raw detector data in quantum cryptosystem, if you know).
I've had to embed several C routines in such places (through the Code Interface VI), and has sometimes been wondering what's the whole point with LabVIEW then...
Another observation (by a colleague of mine) is that he's never seen a program in LabVIEW that looks like a finished application. You'd always have to check first the initial paremeters scattered over different windows and in the source, and then start and stop the program with the Run and Stop buttons on the tool panel. The G language certainly encourages quick-and-dirty programming, even though it is in theory possible to create a standard stand-alone application in it.
And each consumer has the right to decide if it's a ploy or a good offer.
A year ago, spam titles were kinda fun to read.
Today, they're all like hasty business notes from dyslexic employees of a big American company.
The spam war ruined my day.
- Zaputannye sostojanija
- Pereputannye sostojanija
The researchers I met commented that when one talks about this, one picks the translation that best describes his understanding of the subjectI agree. For most sites, it's true, or should I say it's up to them to decide what browsers to support.
There is, however, a class of sites that are mission-critical. Such an important thing as controlling your domain name or finding information on the site of a TLD authority, must work in any browser, down to those used by 0.01% visitors.
Netscape 4.x is still used by perhaps one or two percent visitors, while requiring Flash locks out a hell lot of them.
Besides, not working in Netscape 4, while often being nearly meaningless in itself, is actually a reliable sign of further troubles with your Web design. It's just a correlation that for some reasons exist in today's Web design.
Any TLD authority whose site doesn't work in Netscape 4 looks cheesy to me... (reach for IE) and the one that requires Flash definitely won't be on my list. Period.
Two weeks ago, I read A Plan for Spam article from the last conference, announced on Slashdot. There, the author describes spam-of-the-future as "some completely neutral text followed by a url".
Voila, the future has come. Yesterday I got a short message in Russian, in friendly tone, with an URL. Just like the ones I sometimes get. I'm a webmaster of a site with diverse content, and strangers sometimes send me stuff like this for news etc. There is absolutely no way to tell whether it's a spam or not without visiting the URL.
While the developers wrestle with one strategy and openly discuss the remedies, the spammer sees it and picks the next strategy, always ahead of you! Who benefits more from these conferences, good folks or the spammers?
One fix I'd propose is to stop publishing and webcasting the conference stuff. Then the spammers would have to attend in person. You know what happens next. A spammer surrounded with angry geeks :)
I'm not a space expert, but the most sensible thing for my country (I'm Russian) would be to fully support ISS operation with its Progress and Soyuz spacecrafts, until the things are sorted out with the Shuttles. Perhaps cough up some extra cash on the Russian side, yes. That would also be a politically correct thing to do.
This would mean the construction activity is halted (Shuttles were to deliver most/all new modules), but at least the station can be operated in its current configurations for the time being.
I view the dual delivery systems (STS + Russian crafts) as a partial redundancy built into the ISS program. Don't we now have the exact case when this redundancy should be used?
Any knowledgeable person to comment?
Yes. For one, OpenACS toolkit (and a lot of on-line communities built on it) uses it. TCL is a native language of AOLServer.
This way, it automatically keeps everybody happy until the customer starts overusing the service in a systematic way, which is another story.
Flash is apparently a problem on the Web. Either he excluded it from the list intentionally, or he believes the use of Flash is going to improve radically soon enough, now that he works on it :). Either way, he is biased and doesn't disclose it.
Well good. My browsing improved a lot since I deleted the Flash plug-in. I still don't miss it.
If my university prohibited Napster, as some othes Scandinavian schools did, the collection would probably have never started.
Worse than that, I would never know first-hand what P2P is. This is about academic freedom: you should be allowed to test whatever darn new thing is out there, for whatever reason, otherwise the school lags behind. What you use it for, is your responsibility, of course.
Oh yes, I'm first-hand aware of the associated headaches (cleaning up the lab computers from those pesky money-generating add-ons that pop up an ad at the timing-critical phase of your data acquisition :-).
Here is the original paper on the Nature site. Still, Kurtsiefer's home directory proves interesting as well.
Also, by poking around the (unprotected) Kurtsiefer's home directory, you can have a lot of images of the experiment, experimenters, and the beautiful surroundings.
My congratulations to Dr. Kurtsiefer for finally getting this difficult experiment to work. He's been trying to finish it for about a couple years, if I recall correctly.
This is probably detected as spamming by the search engines. White and/or tiny text has been the pr0n way to boost ranking. You can't count on every search engine to correctly recognize it and index the white text with low priority, instead of ignoring it or, worse yet, firing your entire site from the index.
Without keywords tag, you are left with e.g. this solution (scroll down to the bottom of the page). Not pretty, but search-engine compliant, huh?
Perhaps a better way would be to index these tags with low priority, as some search engines still do. This way, the keywords would only matter if there aren't many other pages with them (misspellings and rare terms), or in conjunction with visible text (variants and attributes). Well, a search engine can check misspelling of common words, but not rare terms and proper names. Both ways, the tags would be hard to abuse while useful in certain searches.
The laziness is working against this (why bother with something which is not visible on the page?), but without meta tags the Web is becoming dummier, in a way. Hope the search engines will master technology to replace them, but it's not quite there yet!
Those who test by FTPing large files and watching the transfer rate, should understand these limitations (kindly explained to me by J.Spencer Love).
I had a similar problem trying to host a large-bandwidth video clip. It turned out the bandwidth of my 10Mbps line did not saturate at all (in fact, it was utilized at mere 5%), so neither did the trans-Atlantic connection. The bottleneck was the internal buffer in client and server software.
This also means you may not need that much bandwidth to push the speed of your FTP/TCP-based tasks to its limit.