As someone who has, in the past, been unfortunate enough to use Adobe's Windows offerings, I'm forced to agree with your assertion that Adobe has no interest in adapting applications to suit the local interface guidelines. Adobe's software has some of the worst UIs I've ever seen, but I suppose that's because I'm not familiar with Mac interfaces.
Mono is just an alternative, I suppose. It might not have any specific advantages, but for someone who has been doing.NET development it provides good crossover and people might prefer the.NET class library to that of Java.
I think trying to implement the 'Windows'* and 'Microsoft' namespaces is a mistake. Firstly, these are not codified in a standard, and are liable to change at any time. In addition, the code and API in the Windows namespace is not designed to be portable, making plenty of assumptions about the Windows registry, Win32 window handles and so forth.
What the Mono people should be doing is making a sturdy, open implementation of the.NET CLR and the 'System' class library, and then let people create libraries for Gtk+ (already done), Qt and anything else. The 'System' class library (to my knowledge) makes no such assumptions about the underlying platform, and a properly-designed application should have the GUI code separate from the main code anyway, so creating a cross-platform application will simply require writing a different implementation of the GUI.
The only loss with this approach is that Mono won't be able to run applications designed for Windows, but I don't consider that a useful goal anyway. What is needed is good, solid open source applications using consistant UI toolkits to create a desktop that feels like it goes together.
* why Microsoft put the 'Windows' namespace under 'System' I have no idea, but let's just pretend they put it under 'Microsoft' where this unportable API logically belongs.
I'd have enough trouble interpreting that render (in the article) if it were made of real objects floating in front of me, but a 2D projection of it would just be hell.
It seems to me that the claim they make about the relationships not being displayable in 2D is false; the parent/child relationships are easy, and we've already got that sorted. The "related by some arbitrary, unspecified characteristic" (grey and yellow folders) can be represented by another pane in the 2D browser for "Things that are related to this elsewhere", which Windows XP already does for lots of its "special folders" as a substitute for actually putting them in a sensible heirarchy in the first place.
A BBC News article on the subject of NOTA votes seems to suggest that "spoiled" (which I take to mean "incorrectly completed") ballot forms are not counted. A few of the comments at the bottom of the article seem to confirm this.
If they were to auto-correct domain typos, they'd have to make the name resolve to an IP of a simple HTTP server which just doles out redirects to the correct domain. If they make it resolve to what they think you meant, cookies will break amongst other things which are tied to domain name. (SSL keys, for example)
For.tv, they've already got this kind of thing, but the server it points at redirects you to a domain purchase page rather than guessing what you meant.
That reminds me of the time I was telling an American friend about someone I knew who had a car which was powered by natural gas. I was talking for several minutes saying that the car was "powered by gas", wondering why my friend was making that "you seem to be talking crap but I'll humour you" face at me. Either that, or it was a "what the hell are British cars normally powered by?" confused look!
The company which runs the Christmas Island country-code domain (cx) has been doing this for a while now. Any unregistered domain within the cx top-level domain will resolve to 219.88.106.80, which then redirects to their site for buying the domain.
Of course, if you put, say, mail.dfjhkhkjdhfgdfgfg.cx into your email client as your POP3 server, you'll end up connecting to 219.88.106.80 on port 110, which thankfully doesn't have anything on the other end of it.
A quick survey reveals that it's also being done for.nu,.tv and.cc. Note that.tv is run by VeriSign, so they've already been trying this - I guess they observed that it worked quite well for getting people to register and have decided to try it for the international domains.
It annoys me that external companies have managed to grab hold of country codes for countries whose ISO codes happen to spell something "cool" in English and started selling them off to organisations in English-speaking countries. It's just plain wrong, damnit.
It's under PC Magazine utilities. I seem to remember you have to register for an account in order to download the software. You have to pay, however.
They bundle the source, so someone could potentially make it work with Windows XP and post a patch. Redistributing anything other than the patch would be disallowed by their licence, which prohibits any kind of redistribution.
Also, the techniques described in the article could potentially be used to reimplement it without using any of their code if anyone wants to.
I have a copy on a PC Magazine coverdisk, which is nice! :)
I hear that a couple of different UK ISPs have been going so far as matching bits of the offending packets so that 'innocent' packets on each port will still get through. This seems like the most ideal solution, but does require more CPU usage on whatever is doing the filtering.
Depending on how granular you want to get, you could (say) block all packets trying to use DCOM, which would shut out legitimate uses of DCOM, or pick out the buffer overflow itself and drop that.
Of course, this is reactive rather than proactive. Worms are only blocked after they've been in the wild long enough to get a fix on what to block, by which time they're probably already in your network and you're perfoming "damage control".
It amuses me when people call RS232 ports "legacy"! However, I guess my claim of all modems talking the same language was partially due to the fact that they operated through relatively a braindead interface. Is there no way to make Windows and/or Linux treat a USB port like an RS232 port and let you just send stuff to a nice, sensible AT-speaking modem?
I don't have much experience with Dell boxes, but most preconfigured-type manufacturers whose products I've "experienced" have shipped with cheap monitors which can only do anything above 800x600 at stupidly low refresh rates (~60Hz).
Is it possible that your X configuration specified a higher refresh rate than the monitor could handle, yet when they ship Windows they tend to have the refresh rate turned down?
One of the PCs I do freebie tech support for (owned by a friend:)) has an on-board modem and Ethernet NIC. However, I believe the modem is a winmodem.
There was a time when modems did speak a common language. One day someone had the bright idea to use glorified soundcards instead of modems and do the modem bit in software. The rest is history.
Sure, I can, aside from the annoyance of my dynamic IP address and having to keep a hostname with a small DNS refresh time up to date.
Your average user doesn't even know what an IP address is, however, let alone how to install a Jabber server. They need geeky friends to show them how or do it for them.
Not everyone has geeky friends. These are the people who use MSN Messenger because they know no better.
I heard about this proposal about a year ago from yet another sensasionalist source. It was quoted as a possibility by someone and made out to be a policy by the less "reliable" members of the media.
This will likely never happen, as there are too many flaws in the plan, and the costs are far too high. The Sun just likes causing a stir.
My house has a wacky mix of screw and bayonet light fittings. Remembering which ones are which and keeping handy a couple of each can be a pain.
It'd be cool if everything used bayonet, but most houses have been around for a while and have "historical" light fittings.
The transition period in North America would be long and most people will see no benefit in changing their existing light fittings, so the screw-sockets would be around for some time to come.
...the deeper the discharge, the more capacity is lost with each cycle...
Does this mean that I shouldn't be letting my phone battery run down as far as possible before I charge it? What is the optimum recharging technique to maximise battery life?
NTL's cable service actually has two TOS agreements. One of them says "do not run any servers whatsoever" and the other says "we don't really care what you run as long as it doesn't cause us any trouble", which I guess means relaying spam and using up lots of download/upload bandwidth would get their attention.
I called them once to clarify and they said it was the second one they actually use, mostly because it's not really worth their effort to police everyone. I suggested that they remove or update the first TOS agreement to avoid confusion. I've no idea if they actually have now, as I can't remember where either of them are!
Slashdot Mirror
It is software being modified, not hardware. They are repairing a buffer overflow valnerability in the dashboard software.
This is quicky becoming the "All your base" of 2003.
As someone who has, in the past, been unfortunate enough to use Adobe's Windows offerings, I'm forced to agree with your assertion that Adobe has no interest in adapting applications to suit the local interface guidelines. Adobe's software has some of the worst UIs I've ever seen, but I suppose that's because I'm not familiar with Mac interfaces.
Mono is just an alternative, I suppose. It might not have any specific advantages, but for someone who has been doing .NET development it provides good crossover and people might prefer the .NET class library to that of Java.
I guess I accidentally hit the "Post Anonymously" checkbox on my way to the preview button. The parent was posted by me.
I think trying to implement the 'Windows'* and 'Microsoft' namespaces is a mistake. Firstly, these are not codified in a standard, and are liable to change at any time. In addition, the code and API in the Windows namespace is not designed to be portable, making plenty of assumptions about the Windows registry, Win32 window handles and so forth.
What the Mono people should be doing is making a sturdy, open implementation of the .NET CLR and the 'System' class library, and then let people create libraries for Gtk+ (already done), Qt and anything else. The 'System' class library (to my knowledge) makes no such assumptions about the underlying platform, and a properly-designed application should have the GUI code separate from the main code anyway, so creating a cross-platform application will simply require writing a different implementation of the GUI.
The only loss with this approach is that Mono won't be able to run applications designed for Windows, but I don't consider that a useful goal anyway. What is needed is good, solid open source applications using consistant UI toolkits to create a desktop that feels like it goes together.
* why Microsoft put the 'Windows' namespace under 'System' I have no idea, but let's just pretend they put it under 'Microsoft' where this unportable API logically belongs.
If it knew which button you wanted.... you wouldn't have to click it!
I'd have enough trouble interpreting that render (in the article) if it were made of real objects floating in front of me, but a 2D projection of it would just be hell.
It seems to me that the claim they make about the relationships not being displayable in 2D is false; the parent/child relationships are easy, and we've already got that sorted. The "related by some arbitrary, unspecified characteristic" (grey and yellow folders) can be represented by another pane in the 2D browser for "Things that are related to this elsewhere", which Windows XP already does for lots of its "special folders" as a substitute for actually putting them in a sensible heirarchy in the first place.
A BBC News article on the subject of NOTA votes seems to suggest that "spoiled" (which I take to mean "incorrectly completed") ballot forms are not counted. A few of the comments at the bottom of the article seem to confirm this.
If they were to auto-correct domain typos, they'd have to make the name resolve to an IP of a simple HTTP server which just doles out redirects to the correct domain. If they make it resolve to what they think you meant, cookies will break amongst other things which are tied to domain name. (SSL keys, for example)
For .tv, they've already got this kind of thing, but the server it points at redirects you to a domain purchase page rather than guessing what you meant.
Not voting is not the answer. What is needed is a "none of the above" vote, which I don't recall seeing on my last voting sheet.
By not voting you increase the weight of the votes by all of the country's idiots.
That reminds me of the time I was telling an American friend about someone I knew who had a car which was powered by natural gas. I was talking for several minutes saying that the car was "powered by gas", wondering why my friend was making that "you seem to be talking crap but I'll humour you" face at me. Either that, or it was a "what the hell are British cars normally powered by?" confused look!
The company which runs the Christmas Island country-code domain (cx) has been doing this for a while now. Any unregistered domain within the cx top-level domain will resolve to 219.88.106.80, which then redirects to their site for buying the domain.
Of course, if you put, say, mail.dfjhkhkjdhfgdfgfg.cx into your email client as your POP3 server, you'll end up connecting to 219.88.106.80 on port 110, which thankfully doesn't have anything on the other end of it.
A quick survey reveals that it's also being done for .nu, .tv and .cc. Note that .tv is run by VeriSign, so they've already been trying this - I guess they observed that it worked quite well for getting people to register and have decided to try it for the international domains.
It annoys me that external companies have managed to grab hold of country codes for countries whose ISO codes happen to spell something "cool" in English and started selling them off to organisations in English-speaking countries. It's just plain wrong, damnit.
It's under PC Magazine utilities. I seem to remember you have to register for an account in order to download the software. You have to pay, however.
They bundle the source, so someone could potentially make it work with Windows XP and post a patch. Redistributing anything other than the patch would be disallowed by their licence, which prohibits any kind of redistribution.
Also, the techniques described in the article could potentially be used to reimplement it without using any of their code if anyone wants to.
I have a copy on a PC Magazine coverdisk, which is nice! :)
I hear that a couple of different UK ISPs have been going so far as matching bits of the offending packets so that 'innocent' packets on each port will still get through. This seems like the most ideal solution, but does require more CPU usage on whatever is doing the filtering.
Depending on how granular you want to get, you could (say) block all packets trying to use DCOM, which would shut out legitimate uses of DCOM, or pick out the buffer overflow itself and drop that.
Of course, this is reactive rather than proactive. Worms are only blocked after they've been in the wild long enough to get a fix on what to block, by which time they're probably already in your network and you're perfoming "damage control".
It amuses me when people call RS232 ports "legacy"! However, I guess my claim of all modems talking the same language was partially due to the fact that they operated through relatively a braindead interface. Is there no way to make Windows and/or Linux treat a USB port like an RS232 port and let you just send stuff to a nice, sensible AT-speaking modem?
I guess USB doesn't really work like that.
I don't have much experience with Dell boxes, but most preconfigured-type manufacturers whose products I've "experienced" have shipped with cheap monitors which can only do anything above 800x600 at stupidly low refresh rates (~60Hz).
Is it possible that your X configuration specified a higher refresh rate than the monitor could handle, yet when they ship Windows they tend to have the refresh rate turned down?
One of the PCs I do freebie tech support for (owned by a friend :)) has an on-board modem and Ethernet NIC. However, I believe the modem is a winmodem.
There was a time when modems did speak a common language. One day someone had the bright idea to use glorified soundcards instead of modems and do the modem bit in software. The rest is history.
If movies do so much better than games, then we should make games based on popular movies.
Therefore I propose that someone make a game which ties in to the popular "Lara Croft: Tomb Raider" film. It should be a big hit!
Sure, I can, aside from the annoyance of my dynamic IP address and having to keep a hostname with a small DNS refresh time up to date.
Your average user doesn't even know what an IP address is, however, let alone how to install a Jabber server. They need geeky friends to show them how or do it for them.
Not everyone has geeky friends. These are the people who use MSN Messenger because they know no better.
Are you going to set up and maintain a Jabber server for all of your friends to use?
Everyone piling off MSN and onto jabber.org or jabber.com is not the answer. For Jabber to work, people must run their own servers.
Centralized messaging sucks, but decentralized IM will never work for the masses unless it's peer-to-peer and "just works".
I heard about this proposal about a year ago from yet another sensasionalist source. It was quoted as a possibility by someone and made out to be a policy by the less "reliable" members of the media.
This will likely never happen, as there are too many flaws in the plan, and the costs are far too high. The Sun just likes causing a stir.
My house has a wacky mix of screw and bayonet light fittings. Remembering which ones are which and keeping handy a couple of each can be a pain.
It'd be cool if everything used bayonet, but most houses have been around for a while and have "historical" light fittings.
The transition period in North America would be long and most people will see no benefit in changing their existing light fittings, so the screw-sockets would be around for some time to come.
"Switch" is short for "switching hub". "Hub" can be used to describe both traditional "dumb" hubs and these new-fangled switching hubs.
Does this mean that I shouldn't be letting my phone battery run down as far as possible before I charge it? What is the optimum recharging technique to maximise battery life?
NTL's cable service actually has two TOS agreements. One of them says "do not run any servers whatsoever" and the other says "we don't really care what you run as long as it doesn't cause us any trouble", which I guess means relaying spam and using up lots of download/upload bandwidth would get their attention.
I called them once to clarify and they said it was the second one they actually use, mostly because it's not really worth their effort to police everyone. I suggested that they remove or update the first TOS agreement to avoid confusion. I've no idea if they actually have now, as I can't remember where either of them are!