But the final output could have done (it looked (almost) identical to the Word version when printed). Their crappy automated submission system was the blocker, not any issue with OOo.
Anyway, OOo seems to me to be no worse than Word for the kind of tasks I used it for.
I'd like to see one person who has made something with OpenOffice that was published professionally.
I've prepared papers for IEEE with OpenOffice. I'd rather have used latex, but my co-author insisted on Word, so I did my share in OO. It worked fine... or at last as well as Word anyway (figures jumping randomly, poor line breaks, etc).
Unfortunately their automated PDF checker b0rks on OO PDFs: you get an unhelpful "there is an error on page 1" message and that's it. So I had to print-to-file from Word using a particular version of some Adobe thing to get a conformant PDF, very annoying.
In Synaptic, click Settings / Repositories, click Add, tick the Universe box, click OK. Now search for WPA again and you should see the package. Except if you don't have a working network connection:-(
You'll also notice more packages available: my Synaptic has 17,000+ of them, heh.
Yes, you're right. For inkjets... it means pretty much zip. It's just a figure produced by marketing departments. Epson used to be very guilty of doing this, maybe they're better now.
Another tip a printer designer told me:) don't think of DPI, think of PPPI, or Pixels Per Printed Inch. Try sending a photo to the printer at higher and higher resolution. At what point do you stop seeing a quality improvement?
For the large format inkjets I used to work with (rated at 600 x 1200 DPI), image quality maxed out at about 150 pppi (because of the size of the dither cell, as you said). You can actually start to see a drop in quality beyond this since the printer is downsizing your image and you'll start to get moire effects. Plus of course your print is taking longer because you're shipping more data to it.
A desktop photo printer will have a much smaller dropsize, so the quality will peak at a higher pppi than that.
A good tip I heard from a printer designer was to ignore the DPI figure, as long as it's more than about 600. It (usually) means how precisely the printer can place dots. It does not say much about the detail or grain you'll see in the print. For that, you need to know the dot size. Of course there's a trade off: smaller dots means (other factors being equal) longer print times, since you have to squirt more dots to get the same level of ink density.
Higher end printers have several shades of grey ink as well as black. This can add a lot of the apparent smoothness of prints, especially if you are going to be printing any black and white photos.
Metamerism is also very important. Print a black and white photo and look at it under tungsten and in daylight. It should stay looking black and white! You'll find some will look red in tungsten and greenish in daylight.
Finally, look at color management. Does the driver let you use your own profiles, or is it more of a point and shoot thing?
HD-DVD is DRM-ed up to the eyeballs as well, there's no real difference. As is the original DVD, of course. HD-DVD may allow playing from hard disc, but it will still be very heavily DRM'd. Blu-ray might also allow this, it's not clear yet.
Heh, actually a caption to a 1940s James Thurber drawing. A host offers a glass to a guest: "It's a naive domestic Burgundy without any breeding, but I think you'll be amused by its presumption."
I don't seem to be able to find a scan of the actual cartoon online anywhere, oh well.
They should glue some valves to the side, use deep cryogenic treatment on the mains cable for the battery charger, and include an integrated magic chip to automatically fix the CDs you copy on to it.
The MS thing is just support for no-execute: the bit that says that this page is only code and not data and you shouldn't try to run anything in here. Everyone has supported this for ages.
This is more. It looks like they are adding extra 'tripwire' pages to the heap, so if an attacker manages to write to part of the heap they shouldn't, there's a good chance they'll hit a tripwire and be detected.
Heh, it did get mangled pretty bad, it's hard to see what's going on there. I've not used gtk_cell_renderer_toggle(), maybe it is very odd like you say. The times I've used other parts of the TreeView widget, it's been fine: a bit wordy, but quite logical and easy to use.
You're right in general that Qt is higher-level than GTK+, since Qt is a C++ API. Of course this has disadvantages too; for example language bindings for Qt are much harder. Gtkmm is worth a look: it's at the same level at Qt (in terms of abstration), but doesn't need MOC. And it plays nice with the STL too.
GtkGLArea sucked quite a bit, it's been replaced by GtkGLExt. Porting is quite easy, certainly less work than gtk1 to gtk2, which you will probably also need to do.
Of course if you don't want to port, you can always run the old version.
That Rpeak is about what the xbox360's CPU is supposed to be capable of. The PS3 CPU can (supposedly) peak at 210 Gflops, so they match the #1 supercomputer from 1995.
Of course real performance is on applications, not just adding up the number of ALUs and multiplying by the clock speed, etc. etc.
That can be true for compact point-and-click cameras with tiny 7mm x 5mm sensors, but not for DSLRs. They have much better dynamic range and lower noise than film.
No DSLR uses multiple CDDs (AFAIK). You'll get rather a good B&W by just taking the green channel.
Finally film resolution is always quoted for some tiny contrast ratio (20%? something like that). Digital resolution is at 100% contrast ratio so it can actually look sharper even when the lpi is lower.
If anyone's not seen it, this DSLR vs medium format shootout from a few years ago has some interesting stuff in. Has a film person made a rebuttal? I'd be interested to see.
I'd add another point: cross-linking of databases. Disclaimer: I'm a brit and I don't know much about the US reasons for wanting national ID, but I imagine the arguments are similiar.
Currently there's no reliable way to check between databases containing personal information. In the UK, for example, we have national insurance numbers, health service numbers, driver's license numbers and so forth, but you can't easily go from one to another. A human looking at the records can make a guess ("Ah ha! This J. Smith born on 1/3/64 is probably this driver and this patient!") but it's not 100% and it's very hard to automate. Plus many of these databases have rather weak checks on who gets a number. It's easy to get a fake health service number, for example.
One of the main reasons the UK govt. is keen on national ID is to fix this problem. If there's a secure link between an individual and a number that can be used anywhere, suddenly all these databases become trivially easy to cross-check. It's such a wonderful thing, that of course private companies will start using it too (if they are allowed).
Summary: governments like ID cards because they make the business of government easier. National ID does nothing (directly) for the citizen, and in fact is harmful to privacy.
At the moment the Mozilla suite has a more recent version of gecko under the hood. The next firefox (1.1 I think it's going to be called, due out in a few months) will be switching to this and get these improvements too. It'll fix some bugs too, eg. the slashdot rendering problem.
This is much higher resolution than the gigapixel people. And digital has better colour than film (though it's not clear how much colour management the Met's photographers did).
Slashdot Mirror
Sadly not ... you have to base64-encode your binary data. I have to do this to get ICC profiles embedded within an XML image file format header :-(
Anyway, OOo seems to me to be no worse than Word for the kind of tasks I used it for.
I've prepared papers for IEEE with OpenOffice. I'd rather have used latex, but my co-author insisted on Word, so I did my share in OO. It worked fine ... or at last as well as Word anyway (figures jumping randomly, poor line breaks, etc).
Unfortunately their automated PDF checker b0rks on OO PDFs: you get an unhelpful "there is an error on page 1" message and that's it. So I had to print-to-file from Word using a particular version of some Adobe thing to get a conformant PDF, very annoying.
void f( int i ) { if( i < 10 ) f( i + 1 ); }
f( 0 );
In Synaptic, click Settings / Repositories, click Add, tick the Universe box, click OK. Now search for WPA again and you should see the package. Except if you don't have a working network connection :-(
You'll also notice more packages available: my Synaptic has 17,000+ of them, heh.
Another tip a printer designer told me :) don't think of DPI, think of PPPI, or Pixels Per Printed Inch. Try sending a photo to the printer at higher and higher resolution. At what point do you stop seeing a quality improvement?
For the large format inkjets I used to work with (rated at 600 x 1200 DPI), image quality maxed out at about 150 pppi (because of the size of the dither cell, as you said). You can actually start to see a drop in quality beyond this since the printer is downsizing your image and you'll start to get moire effects. Plus of course your print is taking longer because you're shipping more data to it.
A desktop photo printer will have a much smaller dropsize, so the quality will peak at a higher pppi than that.
Higher end printers have several shades of grey ink as well as black. This can add a lot of the apparent smoothness of prints, especially if you are going to be printing any black and white photos.
Metamerism is also very important. Print a black and white photo and look at it under tungsten and in daylight. It should stay looking black and white! You'll find some will look red in tungsten and greenish in daylight.
Finally, look at color management. Does the driver let you use your own profiles, or is it more of a point and shoot thing?
HD-DVD is DRM-ed up to the eyeballs as well, there's no real difference. As is the original DVD, of course. HD-DVD may allow playing from hard disc, but it will still be very heavily DRM'd. Blu-ray might also allow this, it's not clear yet.
I don't seem to be able to find a scan of the actual cartoon online anywhere, oh well.
They should glue some valves to the side, use deep cryogenic treatment on the mains cable for the battery charger, and include an integrated magic chip to automatically fix the CDs you copy on to it.
This is more. It looks like they are adding extra 'tripwire' pages to the heap, so if an attacker manages to write to part of the heap they shouldn't, there's a good chance they'll hit a tripwire and be detected.
You're right in general that Qt is higher-level than GTK+, since Qt is a C++ API. Of course this has disadvantages too; for example language bindings for Qt are much harder. Gtkmm is worth a look: it's at the same level at Qt (in terms of abstration), but doesn't need MOC. And it plays nice with the STL too.
Here's a page about it.
Of course if you don't want to port, you can always run the old version.
Have you seen http://home.google.com? Check out this screenshot!
http://maps.google.co.uk/maps?q=cerne+abbas&ll=50. 823269,-2.476473&spn=0.045233,0.057678&t=k&hl=en
That Rpeak is about what the xbox360's CPU is supposed to be capable of. The PS3 CPU can (supposedly) peak at 210 Gflops, so they match the #1 supercomputer from 1995.
Of course real performance is on applications, not just adding up the number of ALUs and multiplying by the clock speed, etc. etc.
Interesting link, thanks.
No DSLR uses multiple CDDs (AFAIK). You'll get rather a good B&W by just taking the green channel.
Finally film resolution is always quoted for some tiny contrast ratio (20%? something like that). Digital resolution is at 100% contrast ratio so it can actually look sharper even when the lpi is lower.
If anyone's not seen it, this DSLR vs medium format shootout from a few years ago has some interesting stuff in. Has a film person made a rebuttal? I'd be interested to see.
Currently there's no reliable way to check between databases containing personal information. In the UK, for example, we have national insurance numbers, health service numbers, driver's license numbers and so forth, but you can't easily go from one to another. A human looking at the records can make a guess ("Ah ha! This J. Smith born on 1/3/64 is probably this driver and this patient!") but it's not 100% and it's very hard to automate. Plus many of these databases have rather weak checks on who gets a number. It's easy to get a fake health service number, for example.
One of the main reasons the UK govt. is keen on national ID is to fix this problem. If there's a secure link between an individual and a number that can be used anywhere, suddenly all these databases become trivially easy to cross-check. It's such a wonderful thing, that of course private companies will start using it too (if they are allowed).
Summary: governments like ID cards because they make the business of government easier. National ID does nothing (directly) for the citizen, and in fact is harmful to privacy.
Ooop, thanks for the correction. Got my moz 1.8 and 1.7 mixed up :-(
At the moment the Mozilla suite has a more recent version of gecko under the hood. The next firefox (1.1 I think it's going to be called, due out in a few months) will be switching to this and get these improvements too. It'll fix some bugs too, eg. the slashdot rendering problem.
This is much higher resolution than the gigapixel people. And digital has better colour than film (though it's not clear how much colour management the Met's photographers did).
I was the orginal author. It's now a GPL project on sourceforge. Check out the javascript client demo (done by Ruven Pillay), very cool.
Ah! Found it.