I resisted for so long because they'd changed a bunch of things I didn't want changed and the new UI was horrible. However, I recently gave in and just made the effort to fix most of the things that annoy me, including installing a set of small icons to replace the huge icons which came with it by default and playing around with the toolbars.
There are still a few little niggly things which bug me (I keep accidentally hitting the drop-down menu on the author/user toggle instead of the button itself because I'm not used to it being there) but I'm sure I'll get used to it all just in time to have to do it all again upgrading to version eight.
I only upgraded from the end of major version six to seven a week or so ago, and they're getting ready for version eight already. I guess I'm destined to be one major version behind everyone else forever.
I get a bit tired of paying again just to get a browser that crashes less. Really, they should roll back bugfixes (but not new features) into older versions; I don't use any of the new features of Opera 7, and only upgraded because of several crashing bugs in Opera 6 that were driving me mad. (both the image loader and the XML parser seemed to have serious problems with large documents, and I'm suspicious that there is a buffer overflow there somewhere.)
I've never used an integrated mail/calendar app. What do you drag and where do you drag it to? I'm having trouble imagining what it would mean to drag mail into a calendar or to drag a day into a mail folder.
Indeed. Somehow UK's Sky One ended up getting the second half of Season 8 before the US, despite the first half showing in the states months before. Episode twelve of both SG-1 and Atlantis are due to air on Sky One on the 4th Jan, which seems to be a week before US viewers will even see episode eleven.
I thought Claudia Black's character was too much like Aeryn, but perhaps that's just because she's a really poor actor? ;)
There is an extension for Thunderbird which adds the calendar capabilities from Sunbird. I use Sunbird as a separate app, so I can't comment on the extent of the integration.
There is also a C++ interface to GDI+, which is in fact just a bunch of header files which define a wrapper around the flat C-style API exposed by gdiplus.dll.
A bunch of Windows XP's standard apps use it, I think. Also, a friend of mine wrote a simple image viewer application using it.
For a long time I used Windows 3.0 and 3.1 without a mouse. The only thing that was made difficult by this was drawing pictures in Paintbrush, but fortunately that didn't come up very often!
Sadly, these days most Windows applications aren't designed with the mouseless in mind. The system tray notification area, for example, is quite hard to drive without the help of the mouse.
For a year or so when I was at University I was awake most during darkness hours. I would get up at about 8pm, go out to the bar or take part in some other "late night" social activity, then from 2am until 9am I had time to persue personal projects, study or whatever before lectures and classes began. In this particular year all of my classes took place before 12pm, which worked out nicely since I could then go to bed and get eight hours sleep before doing it all again.
During the winter months I wasn't seeing much daylight, but I spent the worst of this period at home on Christmas break anyway so my sleep pattern was less constrained. On one day of the week I had a lecture between 11am and 12pm, so I had to eat my main meal of the day at some point between 12am and 2am and then have a snack for supper during said lecture. Fortunately, this particular lecturer didn't mind my eating in his class.
It worked quite well for that year, but over the summer I got a job and had to rotate back to normal, and when I went back to university my schedule was no longer morning-heavy so I had to conform to a more normal schedule. I particularly enjoyed the fact that I could go out to clubs and such and stay out late without feeling like shit for the morning lectures.
I was given a copy of ViaVoice for my birthday one year after complaining about wrist pain. Fortunately, the wrist pain turned out to be due to the poor setup of my chair vs. my table, but I still had a copy of ViaVoice around to toy with.
I was writing small articles and comments with it with little problem; it handled English quite well. However, quite a lot of my typing is to produce code in Perl since that is what I do for a living. Perl has so many funny little symbols that I just spent the entire time trying to figure out what ViaVoice calls different symbols, and once that was sorted having to say long descriptions of what I wanted to type rather than just bashing a button on my keyboard. I gave up in the end.
I don't recommend speech-powered programming to anyone without a specially-targetted language. It would be interesting to try it with COBOL or AppleScript!:)
Perhaps a possible solution to this problem is to use a system like Gnutella or even FreeNet to distribute the tracker advertisement data. A crude solution would be to just "share" the.torrent files on Gnutella, but someone clever could perhaps come up with a better plan that plays off some of the characteristics of BitTorrent and its trackers to make things more efficient. One possible idea that springs to mind is to have the clients for this automatically create and advertise new trackers in preference to the original tracker. This would have the disadvantage that it would split up the pool of available seeds but it would also spread around the liability of running a tracker and make it hard to trace who originally created it. The clients don't even need to be seeding the files they are providing trackers for as long as they can get a good list of sources from a parent tracker, meaning that this could happen automatically with no user intervention and each user wouldn't really know what they were acting as a tracker for, reducing the liability aspect.
I don't really know enough about how BitTorrent works to think about this in any more detail, but I'm sure there are others that could work on this, and probably already are. Leaving the BitTorrent transfer protocol intact assures that it can continue to be used in the current, more efficient way for more "legitimate" transfers, such as retrieving Linux distributions.
Heck, you could even modify URLs to indicate which name resolution scheme it uses.
Someone could set up a set of root servers which represent alternative roots, so you'd have google.com.icann and something.null.opennic. It would require a few special touches due to how the DNS protocol works (Google's nameservers wouldn't know what "google.com.icann" means) but I suspect something could be worked out.
However, I for one wouldn't want to have sites differentiated by something as arbitrary as who is responsible for the root. That's what the root zone we already have is for. Sure, right now the distinction between.com and.net is pretty meaningless, but imagine trying to remember if sites were in.com.icann or.com.mynic.
Unfortunately, as the DNS is structured today the only way to split responsibility is to create additional zones. One entity will always be responsible for managing one zone, and with zones as big as.com and.net ideally we'd want competing zone managers within one zone. Having.com1,.com2 and.com3 would be nasty, but it's too late now to try to split up.com and.net into smaller categories, and users were never very good at dealing with the heirarchy anyway.
Including the root nameserver in a URL isn't much different to just having one root zone where the operator is required to allow other people to play; it'll just add meaningless junk to URLs and make them even harder to share around and remember.
The usage of external Christmas lights is increasing here in the UK as well. When I was a young 'un the biggest public display of Christmas decorations was a string of lights in someone's window, but these days people seem to go all out with giant inflatable Santas, terrible flashing lights and little statues of snowmen and penguins.
The only reason I can see for it is to show off to the rest of the world that you have quite a big disposable income. I don't participate, since I don't have a great deal of it.:)
One of my housemates has a book whose title is "The GIMP". Whenever we have any kind of party or social gathering without fail some newcomer will spot the book, drunkenly stagger over and ask why we have a book about a gimp.
At first it was funny, but it's getting a little tired now.
I believe that the IE component you are thinking of is called "Trident".
Remember that to bring this media player to other platforms they'd also have to port Nullsoft's playback stuff, and Nullsoft isn't exactly famous for writing portable code.
Sometimes lowering the barriers to entry (as technology such as email and the internet do) do more than just let more people get in on it [...]
If I remove the part in parenthesis, the sentence becomes:
Sometimes lowering the barriers to entry do more than just let more people get in on it [...]
The "do" here should almost certainly be "does". When you use parenthesis and similar constructions, be sure that the outlying sentence is able to stand alone. I only bring this up because it tripped me up when I was reading what you had written. Your point is a good one.
People fail to understand that spoken English and written English are, in recent times, different languages. While punctuation originally existed to express the pauses and other markers of speech, today's written language has become optimised for reading. It has been my experience that people who read more have an "optimised" reading technique, and it is these people who are "tripped up" by unusual sentence constructions. Those that read less almost seem as if they are "saying" the words to themselves and "listening" to the result. This is reflected in a person's reading speed, too.
When people tell me they "don't have time" for writing properly, I usually tell them that learning to write in a coherent and "correct" manner takes practice, and the best way to practice is to write well even where it might be unimportant to do so. It also shows a lack of respect for the recipient to send an unstructured email: it forces someone else to spend more time understanding the message than they otherwise would have done.
Indeed. One of my desktop machines here has a Pentium 2 350MHz in it, while my Dell laptop contains a mobile 366MHz chip. Also, the motherboard in the 350MHz box can only take Pentium 2 chips, up to 450MHz.
(It's nice to know that my two main PCs are now officially "old"! Maybe I'll upgrade in a couple of years...)
Of course, if you want to embed figures which actually scale and render properly like the text, you have to screw around trying to get your diagrams in PDF format. In theory that would be easy, but with so few applications going direct to PDF, and with tools that really seem to have a hard time converting EPS to PDF (I've only had GhostScript succeed once, on a really basic document with no text) it becomes quite hellish to author a decent document with pdflatex.
Having said all that, I still use pdflatex regularly, and continue to search for good solutions for producing figures and/or a good (command-line driven) PostScript to PDF conversion tool.
The most obvious reason is that adding a heirarchy reduces the chances of collision. An international non-profit called Foo, a multinational company called Foo and a little firm in Germany can all have the name 'foo' in their respective zones. (foo.org, foo.com, foo.de)
The reason we have the naming problems we do is because there isn't enough heirarchy, combined with the social problem that people have got used to everything being in.com and have trouble remembering any other domain. If little, local companies in the UK had been prevented from getting.com and.net domains from the start then people would be much more accustomed to seeing org.uk, co.uk, com, org, net, net.uk, de, it, se and so on and they'd be better conditioned to remember which TLD belongs to each entity they are committing to memory.
I don't much like the idea of constraining the flat namespace any further by writing off any names which are similar to other names. What happens if I get wind of a new and upcoming company falled Foobarwibblenoo and register fobarwibblenoo.com? They'll then be unable to register their trademark as a domain name without going through me.
I think that is the problem. People do have trouble remembering which site each domain is in. Here in the UK, people get mixed up with.co.uk and.com all the time, which has meant that most UK companies have invested in.com domains to catch the confused. Supermarket chain Iceland even briefly rebranded their company to be called iceland.co.uk even on the realspace shops and bags to try to drive the idea home. I'm sure many people have made the mistake of going to the site about Iceland that lives at Iceland.com at least once. Competitor Tesco just registered both because their name isn't a country. :)
I'm sure that even in the US, most people's exposure to domain names is via advertising, and most organisations advertising are companies. Therefore the most common domain they see is.com, and they just come to assume that everything ends in.com without really knowing what that abbreviation means.
While that's true (for the gtld registries) I still don't like the idea that thousands of otherwise-available domains are pointing at random advertising sites that nobody cares about.
Regardless of whether or not this "should" be done, the fact is that these scripts can register domains much more rapidly than any human can directly -- they just go for anything which has just expired. This would at least partially account for a spike in domain registrations, much as scripts for registering common typos do.
My point was mainly that it's due to the rise of automated registration scripts that registration has spiked.
(Side note: I'd define "squatting" as sitting on any domain providing a "service" that no-one really wanted, preventing a more relevant service from taking that name. Early on this was done on trademarks, and now it's done on expired domains. Either way, the result is lots of domains that can't be used because they are attached to rubbish. If you don't like that, then never mind. The term used doesn't really matter.)
I bet most of this year's domains have been registered by the automated scripts which watch for domain expiry and jump in and register the domain from underneath the owner.
I've seen this happen in no more than a day. It's very annoying, and means people have to move their sites elsewhere and deal with the old site now being at best a page full of adverts and at worst a redirect to some weird porn.
In Durlston Country Park, Dorset, England there is a big stone globe. Sadly, I don't remember it being oriented in the way you are describing, but if someone was to make a similar sculpture oriented in the right direction it would probably last quite a while.
Slashdot Mirror
Wow. That's neat. I didn't realise the buttons worked like that. Thanks!
I resisted for so long because they'd changed a bunch of things I didn't want changed and the new UI was horrible. However, I recently gave in and just made the effort to fix most of the things that annoy me, including installing a set of small icons to replace the huge icons which came with it by default and playing around with the toolbars.
There are still a few little niggly things which bug me (I keep accidentally hitting the drop-down menu on the author/user toggle instead of the button itself because I'm not used to it being there) but I'm sure I'll get used to it all just in time to have to do it all again upgrading to version eight.
I only upgraded from the end of major version six to seven a week or so ago, and they're getting ready for version eight already. I guess I'm destined to be one major version behind everyone else forever.
I get a bit tired of paying again just to get a browser that crashes less. Really, they should roll back bugfixes (but not new features) into older versions; I don't use any of the new features of Opera 7, and only upgraded because of several crashing bugs in Opera 6 that were driving me mad. (both the image loader and the XML parser seemed to have serious problems with large documents, and I'm suspicious that there is a buffer overflow there somewhere.)
I've never used an integrated mail/calendar app. What do you drag and where do you drag it to? I'm having trouble imagining what it would mean to drag mail into a calendar or to drag a day into a mail folder.
Indeed. Somehow UK's Sky One ended up getting the second half of Season 8 before the US, despite the first half showing in the states months before. Episode twelve of both SG-1 and Atlantis are due to air on Sky One on the 4th Jan, which seems to be a week before US viewers will even see episode eleven.
I thought Claudia Black's character was too much like Aeryn, but perhaps that's just because she's a really poor actor? ;)
There is an extension for Thunderbird which adds the calendar capabilities from Sunbird. I use Sunbird as a separate app, so I can't comment on the extent of the integration.
There is also a C++ interface to GDI+, which is in fact just a bunch of header files which define a wrapper around the flat C-style API exposed by gdiplus.dll.
A bunch of Windows XP's standard apps use it, I think. Also, a friend of mine wrote a simple image viewer application using it.
For a long time I used Windows 3.0 and 3.1 without a mouse. The only thing that was made difficult by this was drawing pictures in Paintbrush, but fortunately that didn't come up very often!
Sadly, these days most Windows applications aren't designed with the mouseless in mind. The system tray notification area, for example, is quite hard to drive without the help of the mouse.
For a year or so when I was at University I was awake most during darkness hours. I would get up at about 8pm, go out to the bar or take part in some other "late night" social activity, then from 2am until 9am I had time to persue personal projects, study or whatever before lectures and classes began. In this particular year all of my classes took place before 12pm, which worked out nicely since I could then go to bed and get eight hours sleep before doing it all again.
During the winter months I wasn't seeing much daylight, but I spent the worst of this period at home on Christmas break anyway so my sleep pattern was less constrained. On one day of the week I had a lecture between 11am and 12pm, so I had to eat my main meal of the day at some point between 12am and 2am and then have a snack for supper during said lecture. Fortunately, this particular lecturer didn't mind my eating in his class.
It worked quite well for that year, but over the summer I got a job and had to rotate back to normal, and when I went back to university my schedule was no longer morning-heavy so I had to conform to a more normal schedule. I particularly enjoyed the fact that I could go out to clubs and such and stay out late without feeling like shit for the morning lectures.
I was given a copy of ViaVoice for my birthday one year after complaining about wrist pain. Fortunately, the wrist pain turned out to be due to the poor setup of my chair vs. my table, but I still had a copy of ViaVoice around to toy with.
I was writing small articles and comments with it with little problem; it handled English quite well. However, quite a lot of my typing is to produce code in Perl since that is what I do for a living. Perl has so many funny little symbols that I just spent the entire time trying to figure out what ViaVoice calls different symbols, and once that was sorted having to say long descriptions of what I wanted to type rather than just bashing a button on my keyboard. I gave up in the end.
I don't recommend speech-powered programming to anyone without a specially-targetted language. It would be interesting to try it with COBOL or AppleScript! :)
Perhaps a possible solution to this problem is to use a system like Gnutella or even FreeNet to distribute the tracker advertisement data. A crude solution would be to just "share" the .torrent files on Gnutella, but someone clever could perhaps come up with a better plan that plays off some of the characteristics of BitTorrent and its trackers to make things more efficient. One possible idea that springs to mind is to have the clients for this automatically create and advertise new trackers in preference to the original tracker. This would have the disadvantage that it would split up the pool of available seeds but it would also spread around the liability of running a tracker and make it hard to trace who originally created it. The clients don't even need to be seeding the files they are providing trackers for as long as they can get a good list of sources from a parent tracker, meaning that this could happen automatically with no user intervention and each user wouldn't really know what they were acting as a tracker for, reducing the liability aspect.
I don't really know enough about how BitTorrent works to think about this in any more detail, but I'm sure there are others that could work on this, and probably already are. Leaving the BitTorrent transfer protocol intact assures that it can continue to be used in the current, more efficient way for more "legitimate" transfers, such as retrieving Linux distributions.
Someone could set up a set of root servers which represent alternative roots, so you'd have google.com.icann and something.null.opennic. It would require a few special touches due to how the DNS protocol works (Google's nameservers wouldn't know what "google.com.icann" means) but I suspect something could be worked out.
However, I for one wouldn't want to have sites differentiated by something as arbitrary as who is responsible for the root. That's what the root zone we already have is for. Sure, right now the distinction between .com and .net is pretty meaningless, but imagine trying to remember if sites were in .com.icann or .com.mynic.
Unfortunately, as the DNS is structured today the only way to split responsibility is to create additional zones. One entity will always be responsible for managing one zone, and with zones as big as .com and .net ideally we'd want competing zone managers within one zone. Having .com1, .com2 and .com3 would be nasty, but it's too late now to try to split up .com and .net into smaller categories, and users were never very good at dealing with the heirarchy anyway.
Including the root nameserver in a URL isn't much different to just having one root zone where the operator is required to allow other people to play; it'll just add meaningless junk to URLs and make them even harder to share around and remember.
The usage of external Christmas lights is increasing here in the UK as well. When I was a young 'un the biggest public display of Christmas decorations was a string of lights in someone's window, but these days people seem to go all out with giant inflatable Santas, terrible flashing lights and little statues of snowmen and penguins.
The only reason I can see for it is to show off to the rest of the world that you have quite a big disposable income. I don't participate, since I don't have a great deal of it. :)
One of my housemates has a book whose title is "The GIMP". Whenever we have any kind of party or social gathering without fail some newcomer will spot the book, drunkenly stagger over and ask why we have a book about a gimp.
At first it was funny, but it's getting a little tired now.
I believe that the IE component you are thinking of is called "Trident".
Remember that to bring this media player to other platforms they'd also have to port Nullsoft's playback stuff, and Nullsoft isn't exactly famous for writing portable code.
What we have here is yet another of Netscape's poor design choices (see also IMG, FRAME, FONT) coming back to bite us in the ass.
More grammar nit-picking:
If I remove the part in parenthesis, the sentence becomes:
The "do" here should almost certainly be "does". When you use parenthesis and similar constructions, be sure that the outlying sentence is able to stand alone. I only bring this up because it tripped me up when I was reading what you had written. Your point is a good one.
People fail to understand that spoken English and written English are, in recent times, different languages. While punctuation originally existed to express the pauses and other markers of speech, today's written language has become optimised for reading. It has been my experience that people who read more have an "optimised" reading technique, and it is these people who are "tripped up" by unusual sentence constructions. Those that read less almost seem as if they are "saying" the words to themselves and "listening" to the result. This is reflected in a person's reading speed, too.
When people tell me they "don't have time" for writing properly, I usually tell them that learning to write in a coherent and "correct" manner takes practice, and the best way to practice is to write well even where it might be unimportant to do so. It also shows a lack of respect for the recipient to send an unstructured email: it forces someone else to spend more time understanding the message than they otherwise would have done.
Indeed. One of my desktop machines here has a Pentium 2 350MHz in it, while my Dell laptop contains a mobile 366MHz chip. Also, the motherboard in the 350MHz box can only take Pentium 2 chips, up to 450MHz.
(It's nice to know that my two main PCs are now officially "old"! Maybe I'll upgrade in a couple of years...)
Of course, if you want to embed figures which actually scale and render properly like the text, you have to screw around trying to get your diagrams in PDF format. In theory that would be easy, but with so few applications going direct to PDF, and with tools that really seem to have a hard time converting EPS to PDF (I've only had GhostScript succeed once, on a really basic document with no text) it becomes quite hellish to author a decent document with pdflatex.
Having said all that, I still use pdflatex regularly, and continue to search for good solutions for producing figures and/or a good (command-line driven) PostScript to PDF conversion tool.
The most obvious reason is that adding a heirarchy reduces the chances of collision. An international non-profit called Foo, a multinational company called Foo and a little firm in Germany can all have the name 'foo' in their respective zones. (foo.org, foo.com, foo.de)
The reason we have the naming problems we do is because there isn't enough heirarchy, combined with the social problem that people have got used to everything being in .com and have trouble remembering any other domain. If little, local companies in the UK had been prevented from getting .com and .net domains from the start then people would be much more accustomed to seeing org.uk, co.uk, com, org, net, net.uk, de, it, se and so on and they'd be better conditioned to remember which TLD belongs to each entity they are committing to memory.
I don't much like the idea of constraining the flat namespace any further by writing off any names which are similar to other names. What happens if I get wind of a new and upcoming company falled Foobarwibblenoo and register fobarwibblenoo.com? They'll then be unable to register their trademark as a domain name without going through me.
I think that is the problem. People do have trouble remembering which site each domain is in. Here in the UK, people get mixed up with .co.uk and .com all the time, which has meant that most UK companies have invested in .com domains to catch the confused. Supermarket chain Iceland even briefly rebranded their company to be called iceland.co.uk even on the realspace shops and bags to try to drive the idea home. I'm sure many people have made the mistake of going to the site about Iceland that lives at Iceland.com at least once. Competitor Tesco just registered both because their name isn't a country. :)
I'm sure that even in the US, most people's exposure to domain names is via advertising, and most organisations advertising are companies. Therefore the most common domain they see is .com, and they just come to assume that everything ends in .com without really knowing what that abbreviation means.
While that's true (for the gtld registries) I still don't like the idea that thousands of otherwise-available domains are pointing at random advertising sites that nobody cares about.
Regardless of whether or not this "should" be done, the fact is that these scripts can register domains much more rapidly than any human can directly -- they just go for anything which has just expired. This would at least partially account for a spike in domain registrations, much as scripts for registering common typos do.
My point was mainly that it's due to the rise of automated registration scripts that registration has spiked.
(Side note: I'd define "squatting" as sitting on any domain providing a "service" that no-one really wanted, preventing a more relevant service from taking that name. Early on this was done on trademarks, and now it's done on expired domains. Either way, the result is lots of domains that can't be used because they are attached to rubbish. If you don't like that, then never mind. The term used doesn't really matter.)
I bet most of this year's domains have been registered by the automated scripts which watch for domain expiry and jump in and register the domain from underneath the owner.
I've seen this happen in no more than a day. It's very annoying, and means people have to move their sites elsewhere and deal with the old site now being at best a page full of adverts and at worst a redirect to some weird porn.
In Durlston Country Park, Dorset, England there is a big stone globe. Sadly, I don't remember it being oriented in the way you are describing, but if someone was to make a similar sculpture oriented in the right direction it would probably last quite a while.