Of course I want it all! Isn't that what software promises me?:)
Seriously, I don't think that something needs the huge "markup-to-content" ratio that DocBook has (<para/> for every paragraph? You've got to be kidding me.)
And maybe I should have mentioned that I also want to store structured document markup in a source-code control system, which are most powerful operating on line-oriented text files. And of course, they're all about presentation, not structure.
I do know that nothing is perfect, but it looks like others have made some great suggestions that come closer than anything I've found so far. Thanks for the response!
If you think vi is not easy to use, then you don't know how to use it (probably because, as I pointed out, it's hard to learn).
As far as I'm concerned, as soon as your fingers leave the home position to go arrowing around or (even worse) to go clicking on something, you've already lost the usability battle with vi. It's true that :wq test is much harder to learn than {Move mouse to File, click on File to open the File menu, move mouse to Save As..., click in the filename text entry box, enter the filename, click on the file type selection box to open the file type selection list, click on the scroll arrows to display Plain Text, move mouse to Plain Text, click on Plain Text to select it, move mouse to the Save button, click Save; Move mouse to File, click on File to open the File menu, move mouse to Exit, click on Exit}; but it's easier to use.
[Note: I am aware that some--but only some--of the above steps can be accomplished with keyboard shortcuts. Not all of them, however, and my point remains.]
This guy is confusing usability and learnability. Don't get me wrong--learnability is not a bad goal. The difference is as clear as that between vi and Notepad. There's no question which is more usable, or which Aunt Tillie should use.
Someone else viewing your file won't have the same tab stops and it will be ugly. The file may be viewed with a pager, with cat, formatted for printing with enscript or sent to the printer--all places likely to have tab stops at 8 spaces, which is too much.
Of course these issues are individually soluble (though sending the file to a printer--well, that could be tricky), but the fact is it's a hassle. That's why I don't like tabs.
The fact is, it can sometimes be hard to find the block ends in Python. This isn't a knee-jerk evaluation, it's based on examining and maintaining thousands of lines of a Python application.
I actually do agree that because mixing tabs and spaces is confusing, one tab per indentation level is probably better--but the unfortunate fact is that the code often shows up overly indented in someone else's editor, or when cat-ting or printing. Python proponents (at least the tab flavor) often say, "just set your tab stops to what you like;" but if I have to perform an operation in order to make a file readable I might as well run it through indent which removes the argument to make indentation == syntax in the first place!
I like Python a lot, actually, but I consider this its biggest weakness.
Seriously, really work through it for a couple of weeks, and then one day you will have an epiphany--you will "get it" and become a fan.
I really like Erlang because it's functional programming done in a completely practical, common-sense way. The syntax is elegant without being over-general, and its pattern-matching and function clause selection enable you to write in a really expressive yet compact way.
I haven't tried Haskell because I was originally attracted to Erlang specifically for its concurrency features. Also, I fear and distrust lazy evaluation.
Well (disclaimer: I love VIM) vim is kind of a weird hybrid between the "fully-functional" editor and the "editor-as-tool" philosophies. So it isn't really designed for, say the "search-and-replace in a bunch of files" thing--the idea being that you'd do that in the shell. It does have folding, and you could probably play with the folding patterns to do some of what you want, but I doubt it'd be pretty.
However, why use VIM? I'm sure Emacs can do what you describe, and more besides. If those features are very popular for KEdit I'd be surprised if someone hadn't already implemented them in elisp. If what you want out of VIM is vi-style command keys and moded editing, I'm sure someone's added that to Emacs too.
Well, I'm not saying I think a draft is a good idea, I was only pointing out that I don't think it's the same thing as slavery; as evidenced by comparing the experience of real slaves in this world with those engaged in temporary military service. And I also didn't say the draft was equivalent to paying taxes--in my post I said it was more onerous, didn't I? So you aren't really adding anything by reinforcing the point.
Administering antibiotics is killing organisms, and genocide is killing organisms--so following your simpleminded logic I guess doctors are the same as Hitler, right?
The whole reason to have a draft is so you can pay far below market rates.
Yes, below market, but it depends on your definition of "far." They won't be paying you any less than those of equivalent rank--for a university graduate level specialty it's going to mean at least a warrant officer's billet--looking at the military pay scale (at least for 2002) that's around $25k/yr, a lot more than minimum wage. A general draft for E-1s pays them (again in 2002) $13272/yr, again more than minimum wage. You aren't going to starve.
Conscription is logically equivalent to slavery.
You mean because you can't opt out of it? I don't think this is equivalent to slavery. As citizens, there are several obligations we have to the government, some of them onerous: like taxes. This is just one of them--a particularly onerous one--but since it's temporary and reasonably humane I don't think you can compare it to slavery.
I like Spike but I think the current season is just awful. I'm buying up the previous seasons of Angel as it comes out on DVD and I love the show but I just think the current season is full of misses. Even Spike didn't really hit his groove until recently, his timing and delivery with the rest of the cast were really weird the first few episodes. The Wolfram and Hart storyline just plain sucks; the way that Gunn's character changes is totally out of left field; getting rid of Cordelia cuts the heart out of the show; and for someone who's really loved the way Wesley's character developed over the years, from a bumbling fop to someone with some serious nads (he was the best thing about the last two seasons) to put him back in the slightly goofy researcher role is just slack.
There have been a few good things about the new season--the puppet episode was good; Spike is definitely a bonus (now), but so much of the rest of it's been off that I won't be sorry to see it canceled if renewing would have meant another similar season.
Hopefully Joss is directing his creative energies toward Serenity, the Firefly movie; I admit I can't wait to see it.
The user should be warned that remote printers will not be automatically discovered unless the other machine is so configured.
I like to type.
With that, all of his problems go away. This is a tempest in a teapot, and the tone of the article is unnecessarily mean-spirited: calling the CUPS developers morons, not even half-smart and so forth is uncalled for, when what we're really talking about are three or four details.
Any phone that's connected to the network, even one that has no paid-for service, has the ability to reach 911 at all times.
Yes. And even if you were willing to accept the risk, there are other people in your house at times--workers, or babysitters, the PG&E guy, that may, in an emergency, need to use your phone to dial 911.
This is a pretty good idea--after all, we're going to want it for cellphones, right? This is probably as good as it gets for something like a cellphone (probably better than triangulation), but it seems less useful to me than you might think for a stationary phone.
First, have you used a GPS (especially the tiny, cheap, low-power, less-than-12-channel receivers likely to be in a cellphone or IP phone) inside? GPS can definitely be sketchy. Secondly, sometimes it might be kind of hard to turn a (100-foot-uncertain--more when we're at war) set of coordinates into an understandable street address a police car can drive to: getting a domestic violence call and knowing which block it's on is an order of magnitude less useful than knowing the address.
For static IP addresses, but, consider the variety of ways you could be getting the IP for your VoIP service: cable modem, or DSL--static or dynamic. dynamic PPP over ISDN, etc. It's technically possible to tie, for example, a subscriber's cable modem to an IP address at any given time (probably, but I used to work for a cable modem provider, and you'd be surprised), but there's a lot of integration involved in fitting all the little pieces together so it shows up on a dispatcher's screen.
However, I think such a thing should probably be required, at least at some point. It's usually a requirement for a local police department to respond in some way to every 911 call, even hangups (after all, you could call but be unable to talk).
Then have a ~/.qmail-default file pointing at your real mailbox. Then armorfiend-anything@armorfiend.com will be delivered to your regular mailbox, with the To: header intact. If you start getting spam at one of your disposable addresses, dispose of it by creating a ~/.qmail-amazon with/dev/null in it.
There's a few addresses I can't turn off in this way (like hostmaster@), so I use an opt-in filter for those.
But XP also refers to Extreme Programming; and what programming style could fit Paranoia better than one where your "buddy" looks over your shoulder as you code?
Do you really think that a Chinese government that has the ability to break people's pseudonymity _won't_ do so when dissidents speak out against it?
The same argument can be applied to an anonymous system--that is, freenet's anonymity will be broken by the Chinese government, if it can be, just as with a pseudonymous system.
The entire point of a pseudonym is that it cannot be connected with the poster's, or reader's, real-world identity. But, the pseudonymous identity can be held accountable by community standards (the community of the pseudonymous network), at least to the extent of being silenced (which being the raison d'etre of the network, is effective).
Thanks. It's true, and I looked at (for example) the versions of freenet available for debian-stable. But it would appear from the discussion that even the latest release (which is way newer than Debian) is too old, and the only practical way of using the network was to get the latest nightly build.
Slashdot Mirror
Of course I want it all! Isn't that what software promises me? :)
Seriously, I don't think that something needs the huge "markup-to-content" ratio that DocBook has (<para/> for every paragraph? You've got to be kidding me.)
And maybe I should have mentioned that I also want to store structured document markup in a source-code control system, which are most powerful operating on line-oriented text files. And of course, they're all about presentation, not structure.
I do know that nothing is perfect, but it looks like others have made some great suggestions that come closer than anything I've found so far. Thanks for the response!
If you think vi is not easy to use, then you don't know how to use it (probably because, as I pointed out, it's hard to learn).
As far as I'm concerned, as soon as your fingers leave the home position to go arrowing around or (even worse) to go clicking on something, you've already lost the usability battle with vi. It's true that :wq test is much harder to learn than {Move mouse to File, click on File to open the File menu, move mouse to Save As..., click in the filename text entry box, enter the filename, click on the file type selection box to open the file type selection list, click on the scroll arrows to display Plain Text, move mouse to Plain Text, click on Plain Text to select it, move mouse to the Save button, click Save; Move mouse to File, click on File to open the File menu, move mouse to Exit, click on Exit}; but it's easier to use.
[Note: I am aware that some--but only some--of the above steps can be accomplished with keyboard shortcuts. Not all of them, however, and my point remains.]This guy is confusing usability and learnability. Don't get me wrong--learnability is not a bad goal. The difference is as clear as that between vi and Notepad. There's no question which is more usable, or which Aunt Tillie should use.
Someone else viewing your file won't have the same tab stops and it will be ugly. The file may be viewed with a pager, with cat, formatted for printing with enscript or sent to the printer--all places likely to have tab stops at 8 spaces, which is too much.
Of course these issues are individually soluble (though sending the file to a printer--well, that could be tricky), but the fact is it's a hassle. That's why I don't like tabs.
The fact is, it can sometimes be hard to find the block ends in Python. This isn't a knee-jerk evaluation, it's based on examining and maintaining thousands of lines of a Python application.
I actually do agree that because mixing tabs and spaces is confusing, one tab per indentation level is probably better--but the unfortunate fact is that the code often shows up overly indented in someone else's editor, or when cat-ting or printing. Python proponents (at least the tab flavor) often say, "just set your tab stops to what you like;" but if I have to perform an operation in order to make a file readable I might as well run it through indent which removes the argument to make indentation == syntax in the first place!
I like Python a lot, actually, but I consider this its biggest weakness.
Yea! An Erlang booster!
Seriously, really work through it for a couple of weeks, and then one day you will have an epiphany--you will "get it" and become a fan.
I really like Erlang because it's functional programming done in a completely practical, common-sense way. The syntax is elegant without being over-general, and its pattern-matching and function clause selection enable you to write in a really expressive yet compact way.
I haven't tried Haskell because I was originally attracted to Erlang specifically for its concurrency features. Also, I fear and distrust lazy evaluation.
Well (disclaimer: I love VIM) vim is kind of a weird hybrid between the "fully-functional" editor and the "editor-as-tool" philosophies. So it isn't really designed for, say the "search-and-replace in a bunch of files" thing--the idea being that you'd do that in the shell. It does have folding, and you could probably play with the folding patterns to do some of what you want, but I doubt it'd be pretty.
However, why use VIM? I'm sure Emacs can do what you describe, and more besides. If those features are very popular for KEdit I'd be surprised if someone hadn't already implemented them in elisp. If what you want out of VIM is vi-style command keys and moded editing, I'm sure someone's added that to Emacs too.
I'd heard PL/SQL was from IBM's PL/1, not Ada.
I've been exposed--I give up! jasonditz wins!
I posited a ludicrous false analogy to point out the absurity of the original poster's false analogy--did you not understand that?
Well, I'm not saying I think a draft is a good idea, I was only pointing out that I don't think it's the same thing as slavery; as evidenced by comparing the experience of real slaves in this world with those engaged in temporary military service. And I also didn't say the draft was equivalent to paying taxes--in my post I said it was more onerous, didn't I? So you aren't really adding anything by reinforcing the point.
Administering antibiotics is killing organisms, and genocide is killing organisms--so following your simpleminded logic I guess doctors are the same as Hitler, right?
Yes, below market, but it depends on your definition of "far." They won't be paying you any less than those of equivalent rank--for a university graduate level specialty it's going to mean at least a warrant officer's billet--looking at the military pay scale (at least for 2002) that's around $25k/yr, a lot more than minimum wage. A general draft for E-1s pays them (again in 2002) $13272/yr, again more than minimum wage. You aren't going to starve.
You mean because you can't opt out of it? I don't think this is equivalent to slavery. As citizens, there are several obligations we have to the government, some of them onerous: like taxes. This is just one of them--a particularly onerous one--but since it's temporary and reasonably humane I don't think you can compare it to slavery.
I like Spike but I think the current season is just awful. I'm buying up the previous seasons of Angel as it comes out on DVD and I love the show but I just think the current season is full of misses. Even Spike didn't really hit his groove until recently, his timing and delivery with the rest of the cast were really weird the first few episodes. The Wolfram and Hart storyline just plain sucks; the way that Gunn's character changes is totally out of left field; getting rid of Cordelia cuts the heart out of the show; and for someone who's really loved the way Wesley's character developed over the years, from a bumbling fop to someone with some serious nads (he was the best thing about the last two seasons) to put him back in the slightly goofy researcher role is just slack.
There have been a few good things about the new season--the puppet episode was good; Spike is definitely a bonus (now), but so much of the rest of it's been off that I won't be sorry to see it canceled if renewing would have meant another similar season.
Hopefully Joss is directing his creative energies toward Serenity, the Firefly movie; I admit I can't wait to see it.
Wish I had some mod points. As Dijkstra said:
We are stuck with the term "Computer Science" but would probably better be served by "Computational Science" or something like that.
Note: I'm a programmer, not a computer scientist, and perfectly happy to be one.
What you're actually describing is copylefted software.
Here you're actually describing non-copylefted free software.
Probably the best sources of information are the categories of software listing from the FSF, and the Open Source definition from the OSI.
In practical terms, Open Source is free software, the distinction has to do with the goals and priorities of the respective movements.
No, what he really said:
With that, all of his problems go away. This is a tempest in a teapot, and the tone of the article is unnecessarily mean-spirited: calling the CUPS developers morons, not even half-smart and so forth is uncalled for, when what we're really talking about are three or four details.
Yes. And even if you were willing to accept the risk, there are other people in your house at times--workers, or babysitters, the PG&E guy, that may, in an emergency, need to use your phone to dial 911.
I believe that 911 is the emergency number in good American TV shows as well.
This is a pretty good idea--after all, we're going to want it for cellphones, right? This is probably as good as it gets for something like a cellphone (probably better than triangulation), but it seems less useful to me than you might think for a stationary phone.
First, have you used a GPS (especially the tiny, cheap, low-power, less-than-12-channel receivers likely to be in a cellphone or IP phone) inside? GPS can definitely be sketchy. Secondly, sometimes it might be kind of hard to turn a (100-foot-uncertain--more when we're at war) set of coordinates into an understandable street address a police car can drive to: getting a domestic violence call and knowing which block it's on is an order of magnitude less useful than knowing the address.
For static IP addresses, but, consider the variety of ways you could be getting the IP for your VoIP service: cable modem, or DSL--static or dynamic. dynamic PPP over ISDN, etc. It's technically possible to tie, for example, a subscriber's cable modem to an IP address at any given time (probably, but I used to work for a cable modem provider, and you'd be surprised), but there's a lot of integration involved in fitting all the little pieces together so it shows up on a dispatcher's screen.
However, I think such a thing should probably be required, at least at some point. It's usually a requirement for a local police department to respond in some way to every 911 call, even hangups (after all, you could call but be unable to talk).
Use qmail.
Then have a ~/.qmail-default file pointing at your real mailbox. Then armorfiend-anything@armorfiend.com will be delivered to your regular mailbox, with the To: header intact. If you start getting spam at one of your disposable addresses, dispose of it by creating a ~/.qmail-amazon with /dev/null in it.
There's a few addresses I can't turn off in this way (like hostmaster@), so I use an opt-in filter for those.
But XP also refers to Extreme Programming; and what programming style could fit Paranoia better than one where your "buddy" looks over your shoulder as you code?
The same argument can be applied to an anonymous system--that is, freenet's anonymity will be broken by the Chinese government, if it can be, just as with a pseudonymous system.
The entire point of a pseudonym is that it cannot be connected with the poster's, or reader's, real-world identity. But, the pseudonymous identity can be held accountable by community standards (the community of the pseudonymous network), at least to the extent of being silenced (which being the raison d'etre of the network, is effective).
Thanks. It's true, and I looked at (for example) the versions of freenet available for debian-stable. But it would appear from the discussion that even the latest release (which is way newer than Debian) is too old, and the only practical way of using the network was to get the latest nightly build.