I remember in #linux of EFNet, "man" was the response to about every question.
Those people were wrong. Or at least, only half right. Manpages are a reference source, for situations like "I remember ls can do this, now what's the flag for it?" or "Let's see whether cat has an option for this".
I'm sure all the UNIXes came with documentation for new users (I know AIX has it). For the GNU tools that typically come on a UNIX distribution, the GNU manuals (not manpages) are good. The RedHat manuals are also good.
Then there was also the dreaded "info"...
Yeah, I never clicked with info either -- I assume if you ever managed to learn the Emacs UI, it would be a breeze. Nowadays however, there are GUIs for reading info files, there's info2html, and there's the rather wonderful pinfo.
It'd be even better if you had actual ideas for improving the system, citing examples of other systems that do a better job
Like: "One solution to this might have been to provide a breakdown by first letter, so rather than going "artist -> (long scrolling session) Smiths", I could go "artists by initial -> (short scroll) S -> (short scroll) Smiths"." ?
The fact that you conclude that the iPod is the best available mobile music management system does not make your point a strong one.
I conclude that it's the least-worst HDD based MP3 player. There's no way I would describe iPod as a "mobile music management system". Apple's philospohy appears to be, you manage on iTunes (blech), all you can do on iPod is select and play.
That's a valid design decision, but what irritates me is that part of the appeal of having all those tunes on a piece of random-access mass storage is that I should be able to select a specific track or album pretty much instantly: even at home, it should be more convenient for me to call up an album on iPod than it is to stand up, take a CD off the shelf, put it in the CD player and press play -- yet if you were to time me with a stopwatch, and the artist didn't appear in the first or last 2 screens of the iPod Artists list, then the CD-juggling would win every time.
So you've identified problems that nobody else seems to have much trouble with
My one iPod owning colleague has expressed the same frustrations. I'm sure we're not the only ones. I'm just the only one anal enough to write it all down.
and you think that somebody else should fix them for you.
iPod being a closed platform, only Apple can fix it: although I'll be keeping a close eye on iPod Linux.
Okay. I mean, I guess that's constructive criticism. And I guess you might feel better about yourself assuming that the reason other people don't have this problem is because they're mentally defective.
You're reading between the lines and coming to the wrong conclusions.
Anybody can kvetch about things not working well. It takes a real engineer to make things work better.
Anybody can, but with iPod very few consumers seem to have done: when yet I've identified a number of areas where the iPod UI could do better.
I'm *not* a UI designer: I never could be, my brain doesn't work that way. But likewise, I'm not an automotive engineer, but it wouldn't stop me commenting on the pros and cons of the way my car is designed.
This is assuming that every person has 100% web access, which just isn't true.
I would imagine the main target environment would be the corporate desktop: an instance of the server software would be run internally to the company, so no Internet access is required, just intranet access.
I want to see what the latest additon to the comments are. I only want the previous info in case I need to go back and pick up on a topic I was not involved in form the begining or refresh myself. Otherwise I would rather not scroll the 5000 lines of crap to get the 3 new lines that really matter.
Bottom posting should be combined with vigorous trimming. You should only retain quoted material to which you are directly responding.
Top posting encourages over-quoting, which can be useful in a business environment, but in newsgroups, mailing lists, informal email etc. should definitely be discouraged: if you want the history of a conversation, you should have access to the originals.
"Learning more and more languages/technologies/protocols has merely resulted in a larger skill set on my resume, with pretty much the same level of experience, and no new interviews."
If I was reading a resume and it had an enormous list of languages/protocols, I would be wary of a jack-of-all-trades, master-of-none.
Employers know that if you know the principles of programming and engineering paradigms, then you'll have no problems picking up whatever languages they want you to learn.
So, sure, mention mainstream skills such as C, Perl, Java, HTML, SQL. Don't bother listing Python, Ruby, Miranda, LISP, unless you happen to know that a prospective employer needs one of those. Just say "numerous other programming languages".
Re:"cannot finish understanding vi's modes" ?
on
JOE Hits 3.0
·
· Score: 1
So in that respect, EMACS has one mode less than VI, or two if you count permanent ex and single-line ex (Q and:) as different modes.
I'm not for one moment suggesting that EMACS has lots of modes. Rather I'm jumping on the implication in the article that vi's modes are so numerous that no sane person can get their head around. I'm also keen to rebut the common attitude that modality automatically equals "bad".
I must admit I have great difficulty understanding why people have difficulty with EMACS who claim that they don't with VI. It has a lot of features, but you don't need to use them. People who don't know Meta-X-C will have the same problems as someone in vi who doesn't know:q,:w, etc
Vi was difficult to learn, but the basics you cite (:w to (W)rite,:q to (Q)uit) are at least easy mnemonics.
It sounds easy to remember Meta-X-C to quit (I had to look that up), but when I tried to learn -- and as I pointed out, I have started out with the serious intention of learning EMACS on 3 occasions -- by the time I needed it, I'd forgotten it. One is, after all, introduced to a great many Meta-X-Somethings in the first section of the tutorial.
Is it merely inexperience with EMACS and as a result an assumption that EMACS is "wrong" because they're used to VI?
Possibly. Once you know vi, it quickly becomes invisible -- just as once you can touch-type you don't think about the keyboard, you think about the words, similarly with vi you don't think about vi, you think about the text.
However, while you learn vi, you subconsciously begin to think about text in a different way. The line, the word, the paragraph become units you can move around and between, so when you try to use an editor that doesn't look at text in the same way, it's jarring.
Re:"cannot finish understanding vi's modes" ?
on
JOE Hits 3.0
·
· Score: 1
the ex mode (hit ":")
I didn't count this, because it's akin to saying that (in Windows or Mac) if you've got a pulldown menu open, that's a whole new mode.
Now, there is a full-on ex mode (hit "Q"), but it's not very useful and it's difficult to stumble on, so I glossed over its existence.
"cannot finish understanding vi's modes" ?
on
JOE Hits 3.0
·
· Score: 2, Insightful
"cannot finish understanding vi's modes"?
Give over, man. It has *two* modes! You can always get back to the default mode by hitting escape.
Even iPod has 2 modes (wheel fer scrolling / wheel fer volume)
Now Emacs, that's another matter. I've put serious effort into learning Emacs on three separate occasions in the last 10 years, and every time I gave up because even the simplest thing requires you to remember a seriously obscure series of keystrokes.
-sigh- should have learned not to join editor flamewars by now.
Many groups didn't have good counting systems until very recently. It was common to use concepts like 2 dozen and a three score even two hundred years ago.
You're correct in that even in the last 50 years tribes have been discovered who's numbering systems consist of 1,2 and "more than 2".
But a dozen (12) and a score (20) are perfectly precise, sophisticated numbers that are still in everyday use across the English-speaking world.
With the Mac GUI, the things that the Mac developers decided you might want to do, are convenient.
With scripting as described, pretty much any task you find becoming repetitive can be popped into a script so the computer does the hard work from that point on.
I found that the mini-itx board that I purchased just didn't have the speed to support the application.
The Right Way to do this would be to use dedicated MPEG encoding/decoding hardware -- but you made my point for me. Getting this right needs several iterations of prototyping, by which time you've spent a lot of money on hardware and a lot of effort on tinkering.
Hence I'd buy a Myth based ready-built system, but only when they've got it right.
I keep starting speccing up a mini-itx based PVR, but I always give up because the cost racks up, and that combined with the configuration effort makes sticking with TiVo seem very attractive.
Before I bought one of these I would need:
Some assurance that the system will be whisper quiet (a friend recently put a new drive in his TiVo, and even the noise from that is now irritating)
A case the same width as my other A/V gear
A case-mounted alphanumeric display, so I can navigate MP3s without turning on the TV
The option of at least 2 DVB receivers
1 and 3 are the ones I'm not confident I'd be able to cheaply and easily do on a homebrew box.
Does it actually take 10 hours? We had a provincial election in Ontario a few months back and results were in after an hour or so.
Looks like Ontario has 96 seats to elect.
A UK general election covers 659 seats.
Admittedly my TV viewing pleasure begins an hour or so before the polls close. There follows a race between some of the safer seats, to see who can declare first -- minor errors are acceptable in these seats, because the margins are so large, so they usually come in within the hour. Then results trickle in for the next 5 hours or so, after which it is usually pretty clear which party is in government.
I enjoy making my big cross in a box - but I hate staying up all night to see who won while newsreaders talk shite endlessly.
Really? I love all that. On (UK) election night I get a load of snacks and drinks in, and settle down for 10 hours of exit poll analysis, swingometers, graphs, victories, losses, surprises... I find it tremendously entertaining.
There is a bit of excitement as ministers that you hate get the boot - but overall Id prefer a quicker count.
Ah, I sense that maybe, like me, you cracked open the champagne when Stephen Twigg beat Portillo in 1997. I'd said I'd only open it once there was a genuine majority, but I had to go back on that.
But CompUSA, Best Buy, and other offer "rebated" systems pretty-much exclusively.
Hmm, a (British) friend of mine spent a while in the US working on a white paper. While he was there, he and a German colleague took a trip to Best Buy (or similar) to buy MP3 players. Most came with a mail-in rebate.
The German pointed out to the salesman that the rebate was no good to him, as they could only send the refund to a US resident, so he negotiated a discount in-store.... then he sold the rebate coupon to a US colleage:D
I once used an ATM in Amarillo, TX, which after completing a dial-up referral, would play the first bar of "We're in the money" in monophonic bleep-o-sound (music by Harry Warren, from "Gold Diggers of 1933")
Careful if you're buying a camera phone. They're rather good fun, but you need to know upfront whether you're going to be able to transfer pictures onto a PC directly.
I know a number of people who have to send a costly email/photomessage for every picture they want to move off their phone, because their network operator (from whom they bought the phone) has disabled the functionality to transfer a picture over a wire/bluetooth/IR.
Why don't they just ban cameras? A ban on camera phones doesn't keep someone from sneaking in a small camera. What ignorance.
I just report it, I don't agree with it;)
More to the point, neither a ban on on cameras nor a ban on camera-phones keeps a determined person from sneaking in either. It's not like they have security personnel at the door.
It's knee-jerk pointless policymaking.
(c.f. prohibition of drugs completely failing to solve the problem... but that's offtopic so I'll stop now)
Locker rooms everwhere are in jeapordy, complete with automated pr0n-site deployment!
You joke, but municipal swimming pools in my county have banned camera phones from their changing rooms, for the stated reason that paedophiles might use them.
(Me) I'm no great fan of Java, but I do like the way that there is One True API for most things I might want to do.
(Haeleth) So in the case of Java GUIs, is the One True API Swing or SWT? In case you forgot, there's still an ongoing holy war on that one.
Ah, touche.
I'm going to weasel out of that one by not including GUIS in the set of things that I might want to do: myself being the kind of person who runs away from client side programming and towards servers wherever I can.
Yeah, Java has exceptions to the rule, but in most cases a standard is being settled on... and even for GUIs, Two True APIs, while not as good as One True API, is a better situation than "several dozen APIS: pick one".
This presupposes that the APIs are any good. I'm not a GUI programmer, but people in general seem happy with either Swing or SWT.
I have no faith in these OO language crap either. The real world maybe OO, but once your code is compiled, it is going to run us a sequence of statements: i.e., like an imperative language.
The real world may be OO (the jury's out) but one of the first preconceptions I had to lose was that coding OO was about mapping the Real World to objects. If you look at Design Patterns or similar, you'll find many OO patterns that have no resemblance whatsoever to the way you might naturally try to represent reality.
Slashdot Mirror
I remember in #linux of EFNet, "man" was the response to about every question.
Those people were wrong. Or at least, only half right. Manpages are a reference source, for situations like "I remember ls can do this, now what's the flag for it?" or "Let's see whether cat has an option for this".
I'm sure all the UNIXes came with documentation for new users (I know AIX has it). For the GNU tools that typically come on a UNIX distribution, the GNU manuals (not manpages) are good. The RedHat manuals are also good.
Then there was also the dreaded "info"...
Yeah, I never clicked with info either -- I assume if you ever managed to learn the Emacs UI, it would be a breeze. Nowadays however, there are GUIs for reading info files, there's info2html, and there's the rather wonderful pinfo.
It'd be even better if you had actual ideas for improving the system, citing examples of other systems that do a better job
Like: "One solution to this might have been to provide a breakdown by first letter, so rather than going "artist -> (long scrolling session) Smiths", I could go "artists by initial -> (short scroll) S -> (short scroll) Smiths"." ?
The fact that you conclude that the iPod is the best available mobile music management system does not make your point a strong one.
I conclude that it's the least-worst HDD based MP3 player. There's no way I would describe iPod as a "mobile music management system". Apple's philospohy appears to be, you manage on iTunes (blech), all you can do on iPod is select and play.
That's a valid design decision, but what irritates me is that part of the appeal of having all those tunes on a piece of random-access mass storage is that I should be able to select a specific track or album pretty much instantly: even at home, it should be more convenient for me to call up an album on iPod than it is to stand up, take a CD off the shelf, put it in the CD player and press play -- yet if you were to time me with a stopwatch, and the artist didn't appear in the first or last 2 screens of the iPod Artists list, then the CD-juggling would win every time.
So you've identified problems that nobody else seems to have much trouble with
My one iPod owning colleague has expressed the same frustrations. I'm sure we're not the only ones. I'm just the only one anal enough to write it all down.
and you think that somebody else should fix them for you.
iPod being a closed platform, only Apple can fix it: although I'll be keeping a close eye on iPod Linux.
Okay. I mean, I guess that's constructive criticism. And I guess you might feel better about yourself assuming that the reason other people don't have this problem is because they're mentally defective.
You're reading between the lines and coming to the wrong conclusions.
Anybody can kvetch about things not working well. It takes a real engineer to make things work better.
Anybody can, but with iPod very few consumers seem to have done: when yet I've identified a number of areas where the iPod UI could do better.
I'm *not* a UI designer: I never could be, my brain doesn't work that way. But likewise, I'm not an automotive engineer, but it wouldn't stop me commenting on the pros and cons of the way my car is designed.
25+ buttons? Do they not know that people like(d) the ipod for ease of use, amongst other things?
I've thought a lot about the iPod UI, and it's neither as easy nor as powerful as I would have liked. I wrote this about it.
Most of what I'd like fixed would be easier to do with a couple more buttons. 25 does seem a little excessive however...
This is assuming that every person has 100% web access, which just isn't true.
I would imagine the main target environment would be the corporate desktop: an instance of the server software would be run internally to the company, so no Internet access is required, just intranet access.
I want to see what the latest additon to the comments are. I only want the previous info in case I need to go back and pick up on a topic I was not involved in form the begining or refresh myself. Otherwise I would rather not scroll the 5000 lines of crap to get the 3 new lines that really matter.
Bottom posting should be combined with vigorous trimming. You should only retain quoted material to which you are directly responding.
Top posting encourages over-quoting, which can be useful in a business environment, but in newsgroups, mailing lists, informal email etc. should definitely be discouraged: if you want the history of a conversation, you should have access to the originals.
"Learning more and more languages/technologies/protocols has merely resulted in a larger skill set on my resume, with pretty much the same level of experience, and no new interviews."
If I was reading a resume and it had an enormous list of languages/protocols, I would be wary of a jack-of-all-trades, master-of-none.
Employers know that if you know the principles of programming and engineering paradigms, then you'll have no problems picking up whatever languages they want you to learn.
So, sure, mention mainstream skills such as C, Perl, Java, HTML, SQL. Don't bother listing Python, Ruby, Miranda, LISP, unless you happen to know that a prospective employer needs one of those. Just say "numerous other programming languages".
So in that respect, EMACS has one mode less than VI, or two if you count permanent ex and single-line ex (Q and :) as different modes.
:q, :w, etc
:q to (Q)uit) are at least easy mnemonics.
I'm not for one moment suggesting that EMACS has lots of modes. Rather I'm jumping on the implication in the article that vi's modes are so numerous that no sane person can get their head around. I'm also keen to rebut the common attitude that modality automatically equals "bad".
I must admit I have great difficulty understanding why people have difficulty with EMACS who claim that they don't with VI. It has a lot of features, but you don't need to use them. People who don't know Meta-X-C will have the same problems as someone in vi who doesn't know
Vi was difficult to learn, but the basics you cite (:w to (W)rite,
It sounds easy to remember Meta-X-C to quit (I had to look that up), but when I tried to learn -- and as I pointed out, I have started out with the serious intention of learning EMACS on 3 occasions -- by the time I needed it, I'd forgotten it. One is, after all, introduced to a great many Meta-X-Somethings in the first section of the tutorial.
Is it merely inexperience with EMACS and as a result an assumption that EMACS is "wrong" because they're used to VI?
Possibly. Once you know vi, it quickly becomes invisible -- just as once you can touch-type you don't think about the keyboard, you think about the words, similarly with vi you don't think about vi, you think about the text.
However, while you learn vi, you subconsciously begin to think about text in a different way. The line, the word, the paragraph become units you can move around and between, so when you try to use an editor that doesn't look at text in the same way, it's jarring.
the ex mode (hit ":")
I didn't count this, because it's akin to saying that (in Windows or Mac) if you've got a pulldown menu open, that's a whole new mode.
Now, there is a full-on ex mode (hit "Q"), but it's not very useful and it's difficult to stumble on, so I glossed over its existence.
"cannot finish understanding vi's modes"?
Give over, man. It has *two* modes! You can always get back to the default mode by hitting escape.
Even iPod has 2 modes (wheel fer scrolling / wheel fer volume)
Now Emacs, that's another matter. I've put serious effort into learning Emacs on three separate occasions in the last 10 years, and every time I gave up because even the simplest thing requires you to remember a seriously obscure series of keystrokes.
-sigh- should have learned not to join editor flamewars by now.
Many groups didn't have good counting systems until very recently. It was common to use concepts like 2 dozen and a three score even two hundred years ago.
You're correct in that even in the last 50 years tribes have been discovered who's numbering systems consist of 1,2 and "more than 2".
But a dozen (12) and a score (20) are perfectly precise, sophisticated numbers that are still in everyday use across the English-speaking world.
You miss the point.
With the Mac GUI, the things that the Mac developers decided you might want to do, are convenient.
With scripting as described, pretty much any task you find becoming repetitive can be popped into a script so the computer does the hard work from that point on.
I found that the mini-itx board that I purchased just didn't have the speed to support the application.
The Right Way to do this would be to use dedicated MPEG encoding/decoding hardware -- but you made my point for me. Getting this right needs several iterations of prototyping, by which time you've spent a lot of money on hardware and a lot of effort on tinkering.
Hence I'd buy a Myth based ready-built system, but only when they've got it right.
Before I bought one of these I would need:
1 and 3 are the ones I'm not confident I'd be able to cheaply and easily do on a homebrew box.
Does it actually take 10 hours? We had a provincial election in Ontario a few months back and results were in after an hour or so.
Looks like Ontario has 96 seats to elect.
A UK general election covers 659 seats.
Admittedly my TV viewing pleasure begins an hour or so before the polls close. There follows a race between some of the safer seats, to see who can declare first -- minor errors are acceptable in these seats, because the margins are so large, so they usually come in within the hour. Then results trickle in for the next 5 hours or so, after which it is usually pretty clear which party is in government.
Some seats don't declare until the following day.
I enjoy making my big cross in a box - but I hate staying up all night to see who won while newsreaders talk shite endlessly.
Really? I love all that. On (UK) election night I get a load of snacks and drinks in, and settle down for 10 hours of exit poll analysis, swingometers, graphs, victories, losses, surprises... I find it tremendously entertaining.
There is a bit of excitement as ministers that you hate get the boot - but overall Id prefer a quicker count.
Ah, I sense that maybe, like me, you cracked open the champagne when Stephen Twigg beat Portillo in 1997. I'd said I'd only open it once there was a genuine majority, but I had to go back on that.
But CompUSA, Best Buy, and other offer "rebated" systems pretty-much exclusively.
... then he sold the rebate coupon to a US colleage :D
Hmm, a (British) friend of mine spent a while in the US working on a white paper. While he was there, he and a German colleague took a trip to Best Buy (or similar) to buy MP3 players. Most came with a mail-in rebate.
The German pointed out to the salesman that the rebate was no good to him, as they could only send the refund to a US resident, so he negotiated a discount in-store.
I once used an ATM in Amarillo, TX, which after completing a dial-up referral, would play the first bar of "We're in the money" in monophonic bleep-o-sound (music by Harry Warren, from "Gold Diggers of 1933")
Careful if you're buying a camera phone. They're rather good fun, but you need to know upfront whether you're going to be able to transfer pictures onto a PC directly.
I know a number of people who have to send a costly email/photomessage for every picture they want to move off their phone, because their network operator (from whom they bought the phone) has disabled the functionality to transfer a picture over a wire/bluetooth/IR.
Why don't they just ban cameras? A ban on camera phones doesn't keep someone from sneaking in a small camera. What ignorance.
;)
I just report it, I don't agree with it
More to the point, neither a ban on on cameras nor a ban on camera-phones keeps a determined person from sneaking in either. It's not like they have security personnel at the door.
It's knee-jerk pointless policymaking.
(c.f. prohibition of drugs completely failing to solve the problem... but that's offtopic so I'll stop now)
Locker rooms everwhere are in jeapordy, complete with automated pr0n-site deployment!
You joke, but municipal swimming pools in my county have banned camera phones from their changing rooms, for the stated reason that paedophiles might use them.
I just can't help but think it looks like Crimson Skies with giant robots
Crimson Skies with giant robots is... Crimson Skies: High Road to Revenge.
[Minor Spoiler] A couple of times in the game you take on a giant 8-legged robot: once among the canyons of Atrixo, and once in stormy Chicago.
(Me)
I'm no great fan of Java, but I do like the way that there is One True API for most things I might want to do.
(Haeleth)
So in the case of Java GUIs, is the One True API Swing or SWT? In case you forgot, there's still an ongoing holy war on that one.
Ah, touche.
I'm going to weasel out of that one by not including GUIS in the set of things that I might want to do: myself being the kind of person who runs away from client side programming and towards servers wherever I can.
Yeah, Java has exceptions to the rule, but in most cases a standard is being settled on... and even for GUIs, Two True APIs, while not as good as One True API, is a better situation than "several dozen APIS: pick one".
This presupposes that the APIs are any good. I'm not a GUI programmer, but people in general seem happy with either Swing or SWT.
I have no faith in these OO language crap either. The real world maybe OO, but once your code is compiled, it is going to run us a sequence of statements: i.e., like an imperative language.
The real world may be OO (the jury's out) but one of the first preconceptions I had to lose was that coding OO was about mapping the Real World to objects. If you look at Design Patterns or similar, you'll find many OO patterns that have no resemblance whatsoever to the way you might naturally try to represent reality.