What programs, if any, do those of you reading/. use to keep track of your books? If you were to start an open source internet book database project, what features would you include in it?
What purpose would such a database serve? CDDB/freedb, for example, allow us to automatically download the album titles automatically. Saves everyone a lot of tedious work. Obviously, you're not going to be doing this for books.
As a graduate student, I maintain a single text file of all articles and texts that I've ever referenced. Each entry has a unique identified which I use the UIDs in my own articles instead of typing the full reference. A shell script then updates then updates the references and BibTeX automatically generates the bibliography.
I could see where it could be useful to have a centralized resource that could automatically download those references - but only if it was quicker/easier than typing it in myself (and that only takes a couple of seconds).
What other purposes would such a database serve? How would it make my life easier?
But they claim to have accurate results when they've tested 55 kids to represent how tens of millions of kids globally will use the web? This is balderdash, I say! They did not take enough samples. They should go test several THOUSAND children and them come back with results.
I mean, would you trust a study that calims to provide the innermost secrets of online behaviour of the average american adult when they have observed only 55 people in the whole country? I doubt it.
It's actually worse than what you suggest since the respondents probably weren't randomly selected. With random selection, you only need a handful of respondents to get a good picture of the population (for example, just a couple of thousand randomly selected respondents will give you a great description of the U.S.). But they probably just used people that they knew or were available (i.e. an after-school group or some such) - no better than an online poll ("45% of kids couldn't figure out how to use CowboyNeal to navigate the web.")
I know that Nielsen goes on and on about how you only need 5 or 10 respondents to identify usability patterns. And to a certain extent I think he's right. But there he's talking about adults who have already been socialized in the "computer culture" (for lack of a better term). But these are kids! They're still learning how to use the computer and surf the web. Adults (for better or for worse) have already learned how to use the web - kids are still learning. That completely changes the rules. What may be completely unintuitive to some may be (pun intended) child's play to others.
Boys were significantly more annoyed by verbose pages than were girls (40% of the boys complained, compared to 8% of the girls), possibly because at the ages we tested, boys are not as accomplished at reading as girls.
If I had Jakob Nielsen (most likely shouting "Micropayments are the answer! Micropayments are the answer!") hovering over me as I tried to navigate the web, I'd complain too!
You've only got two days and from what it sounds like, students are being forced to attend so they're not going to want to be there. They're not going to learn much unless you peak their interest.
I'd suggest teaching them the basics of the Unix philosophy - small is beautiful, make each program do one thing well, avoid captive user interfaces, etc. If you don't already have a copy, pick up Gancarz's The Unix Philosophy which describes the various tenets clearly and concisely. (Hell, if you're allowed to or have a budget, make the text a required purchase. It's cheap.)
Something like this would be far more useful than 'ls|grep' or "/etc is where system-wide configuration files are stored" because it would provide them with a new way of approaching computers.
p.s. For extra points, contrast the Unix philosophy which assumes that you know what you're doing with the DOS/Windows philosophy which assumes that you don't know what you're doing. You might even want to begin your class by reading/assigning Scoville's Unix as Literature which nicely depicts how different Unix is from other operating systems.
I don't know if it's related, but I've seen people with electronic agendas that produce an electronic beep each time they type a key. Once in a bus and once in a library, very annoying to others. I wish they would turn the feature off, but these non-geek users are probably thinking "it has to beep, it's electronic".
I certainly agree with this sentiment. For what it's worth, I think that it's a matter of finding the right balance. My Palm Pilot has a very unobtrusive click that I find very useful (for example, if I scroll accidentally); on the other hand, my cell phone has a louder beep that I often find distracting (however, I'm not often typing more than a few numbers on my cell so it's not as bad.)
Besides, if you've ever used a computer while listening to music with headphones, then you've been using a computer without hearings the clicks. Did you miss them? Audio feedback might not be as necessary as you think.
Actually, even though I almost always listen to music when on the computer, I almost never use headphones. (If I'm in a public place, I simply won't listen to headphones.) However, I'm not sure if this is actually relevant to the current discussion. Mostly, I think it's because I don't like being disconnected from my environment. In any case, it means that I haven't experienced what you're talking about.
The few times that I have used headphones while on the computer was for gaming (back when I used to have time to game). But in first person shooters, the game itself provides audio feedback so I never felt that I was missing anything.
Finally, I guess that I should say that I don't think that audio feedback is absolutely necessary. If that were the case, deaf people wouldn't be able to type. However, I do feel that multiple forms of feedback (when, as you point out, it is non-obtrusive) is advantageous.
Not to troll, but why do you want a no-click mouse in the first place? The clicking sound is audio feedback that improves usability. Same reason that keyboards click. More feedback (tactile, visual, audio) is more useful.
It's one thing if you're asking out of curiosity or just to see if you can do it. But if you think that there's a real, measurable benefit to a silent-click, I think that you might want to spend some time reconsidering.
I guess the general question would be: "Under what circumstances would having a silent-click be beneficial?"
I'm looking forward to it. I could never make any progress in Myst because I'm color-blind. Switching indicators from red to green? I don't think so. (Took me days to figure out what I was supposed to be trying to do.) Now I'll finally be able to figure out what the hell was going on!
Can some of the clever folks here at/. come up with a way for Open Source to succeed and pay programmers at the same time?
Nobody needs to come up with a "clever way;" it's already been done. The vast majority of software is produced under contract or for internal-use only.
This is often precisely the types of software that should be open-sourced: software that is responsible for a company's infrastructure but does not provide a competitive advantage. If a particular piece of software is that important to a company's business plan, I can understand why they would want to keep it to themselves. In most cases, however, the company would be better served by using (and contributing to) community-developed software - software that has been well tested, is extremely stable, and very inexpensive.
You don't need to produced shrinkwrapped software in order to make money. Rather, programmers should sell (rent) their services. In fact, to my mind, the idea that we wait for someone to produce the software that we want is ass-backwards. If there's a piece of software that I need, and I can't code it myself, I should just go hire someone to write it for me. (And, when I get back into grad school and have grant money available to me - that's precisely what I intend to do.)
I have the same problem. A few years ago, my [ex-]employer forced me to get an MSN IM account. Just recently, I've begun receiving unsolicited email from Microsoft at that address. What's particulary frustrating is that there is no way to unsubscribe from Passport!
Joel Spolsky argues that offices are essential for developers because it's the only way to get into "the zone." Without a door to close, cubicles invite interruption. Somebody innocently asks a question and bang! all of the variables, procedures, and methods drop out of your head. Takes a good 5, 10, 15 minutes or more to get back to find your place. Considering how much programmers are paid, that adds up to serious money.
Or, like I just did, you might just start reading Slashdot. I gotta get back to work...
Actually, although his company, FogCreek Software, sells proprietary software, they do provide the customer with a copy of the source-code for some (all?) of their products. It's not Open Source (and certainly not Free Software) but you can review the code yourself, fix bugs, etc...
It seems to be very much a YMMV type of thing. Also, it very much depends on what you're talking about. Mozilla's always been slow at starting up. (Preloading dlls does help things for Windows; also, under Windows, I install only the browser. I don't know if that matters or not.)
Regarding page rendering, I've found builds 0.9.8 and 0.9.9 to be incredibly fast. Even Slashdot! In fact, Mozilla renders pages significantly faster on my Pentium 200 Debian Potato system than IE does on my Pentium III 500 Win2K system. (I don't know if it's significant but for my Debian system, I use the prebuilt tar.gz package that doesn't include the Talkback app. Mozilla hasn't crashed on me in quite some time but I've had bad experiences with the Talkback builds in the past.)
I'm really not trying to troll here but anti-aliased text has always looked fuzzy to me. In the screen shots, for example, the spacing and sizing of the AA text is certainly nicer but the default text seems shaper and crisper. Am I wrong here? Joel Spolsky agrees with me but everyone else seems so excited about AA text that I have to wonder if I'm missing something.
I thought that this was great until I read the actual laws involved. The California Business & Professions Code 17538.45 (the law that the individual sued under) states that an Email Service Provider (ESP) may sue individuals and companies that send Unsolicited Commercial Email (UCE) over their servers.
The only reason that the individual was able to win her suit was because her email is forwarded to one machine to another which (in her words) 'may make [her] qualify as an electronic mail service provider'.
This is indeed a victory but it's a small one. The responsibility still lies with the ISPs. Most individuals won't qualify as an ESP and, therefore, can't sue.
The PCGuide at www.pcguide.com has a wonderful discussion of backup and disaster recovery procedures at http://www.pcguide.com/care/bu/index.htm. Discusses the various methods and mediums, rotation schedules, onsite vs offsite storage, etc. Also has an analysis of the how well various backup methods protect against various risks (i.e., hardware failure, file system corruption, accidental deletion). A must read.
As a graduate student in sociology, computing and programming is central to my work. This doesn't mean, however, that I practice computer-science. Computers are simply the tool which I used to conduct my work.
In my experience, most academics in the natural and social sciences are fairly good programmers. Among social scientists, we've all had to learn a variety of program-specific scripting languages (e.g. SPSS and SAS) in order to conduct our analyses. But few of us would count as (or consider ourselves to be) "hackers" or "compter scientists". And, while I've collected a number of programming languages over the years, they are simply the various tools that I use to collect and analyze data. Sed and awk are indispensible and Python has proved quite useful. I write all my papers in LaTeX and expect that I'll eventually learn enough Lisp to tame Emacs. But I'll probably never need C.
Even my father, who, as a graduate student at Stanford during the 60's, was writing Fortran programs in order to conduct his statistical analysis, has never considered himself a programmer. Fortran, too, was simply a tool.
We don't care -- or need to care -- about the theory underlying computing.
With all of the above being said, there is a current development within the social scientific academic community that cast further doubt upon the article's premise -- and one that I find rather worrisome. Rather than increasing, the programming abilities of the academic community (at least within the social sciences) may be declining. And, I believe, that this could have very serious consequences.
Being raised on GUIs, current graduate students are being taught to point-and-click their way through analysis. SPSS, for example, has, over the years, built up a rather nice nice GUI front-end. But, as in all GUIs, the SPSS front-end is less powerful and less flexible than the underlying scripting language.
But students aren't being taught the scripting languages. As a result, many students don't know what their programs are capable of or how to conduct complex analysis. I suspect that many students don't even realize that they can conduct more complex analysis! And, of course, most of the programs used to conduct complex analysis are entirely text based. What I find most worrisome, however, is that the GUIs serve to insulate the user from the analytical process -- as a result, students may be failing to understand and appreciate the statistical theory underlying their analysis.
As social scientists, we are not computer scientists; we don't need to understand computer theory. We do, however, need to understand statistical theory.
Re:Another warning against Linux certification
on
Linuxgruven Deorbits
·
· Score: 1
Actually, anyone who works professionally in Linux should be in favor of certifications. The entire point of certifications (and professional organizations) is to reduce the number of "qualified" applicants so that those who have the certifications can charge more for their services.
This isn't to say that certification procedures aren't crap and aren't meaningless -- but to focus upon that piece of the puzzle is to miss to point of what certifictions are really about.
Why do employees always (usually) believe that managers are assholes? For the same reason that managers always (usually) think that employees are lazy. It's inherent in the relationship.
The entire reason for management's existence is to increase the productivity of workers. Productivity can be increased in only two ways -- by inducing workers to work harder or by forcing workers to work harder.
We in the IT industry generally have it pretty good -- inducement is far more pleasurable than domination. But domination always comes into play at some point, somewhere. Furthermore, the threat of domination is omni-present. And that puts great strain on the relationship.
Slashdot Mirror
What programs, if any, do those of you reading /. use to keep track of your books? If you were to start an open source internet book database project, what features would you include in it?
What purpose would such a database serve? CDDB/freedb, for example, allow us to automatically download the album titles automatically. Saves everyone a lot of tedious work. Obviously, you're not going to be doing this for books.
As a graduate student, I maintain a single text file of all articles and texts that I've ever referenced. Each entry has a unique identified which I use the UIDs in my own articles instead of typing the full reference. A shell script then updates then updates the references and BibTeX automatically generates the bibliography.
I could see where it could be useful to have a centralized resource that could automatically download those references - but only if it was quicker/easier than typing it in myself (and that only takes a couple of seconds).
What other purposes would such a database serve? How would it make my life easier?
But they claim to have accurate results when they've tested 55 kids to represent how tens of millions of kids globally will use the web? This is balderdash, I say! They did not take enough samples. They should go test several THOUSAND children and them come back with results.
I mean, would you trust a study that calims to provide the innermost secrets of online behaviour of the average american adult when they have observed only 55 people in the whole country? I doubt it.
It's actually worse than what you suggest since the respondents probably weren't randomly selected. With random selection, you only need a handful of respondents to get a good picture of the population (for example, just a couple of thousand randomly selected respondents will give you a great description of the U.S.). But they probably just used people that they knew or were available (i.e. an after-school group or some such) - no better than an online poll ("45% of kids couldn't figure out how to use CowboyNeal to navigate the web.")
I know that Nielsen goes on and on about how you only need 5 or 10 respondents to identify usability patterns. And to a certain extent I think he's right. But there he's talking about adults who have already been socialized in the "computer culture" (for lack of a better term). But these are kids! They're still learning how to use the computer and surf the web. Adults (for better or for worse) have already learned how to use the web - kids are still learning. That completely changes the rules. What may be completely unintuitive to some may be (pun intended) child's play to others.
Boys were significantly more annoyed by verbose pages than were girls (40% of the boys complained, compared to 8% of the girls), possibly because at the ages we tested, boys are not as accomplished at reading as girls.
If I had Jakob Nielsen (most likely shouting "Micropayments are the answer! Micropayments are the answer!") hovering over me as I tried to navigate the web, I'd complain too!
D'oh! It's "pique" not "peak"...
You've only got two days and from what it sounds like, students are being forced to attend so they're not going to want to be there. They're not going to learn much unless you peak their interest.
I'd suggest teaching them the basics of the Unix philosophy - small is beautiful, make each program do one thing well, avoid captive user interfaces, etc. If you don't already have a copy, pick up Gancarz's The Unix Philosophy which describes the various tenets clearly and concisely. (Hell, if you're allowed to or have a budget, make the text a required purchase. It's cheap.)
Something like this would be far more useful than 'ls|grep' or "/etc is where system-wide configuration files are stored" because it would provide them with a new way of approaching computers.
p.s. For extra points, contrast the Unix philosophy which assumes that you know what you're doing with the DOS/Windows philosophy which assumes that you don't know what you're doing. You might even want to begin your class by reading/assigning Scoville's Unix as Literature which nicely depicts how different Unix is from other operating systems.
(If I'm in a public place, I simply won't listen to headphones.)
Check that... If I'm in a publick place, I simply won't listen to music.
And I previewed and everything... sigh...
I don't know if it's related, but I've seen people with electronic agendas that produce an electronic beep each time they type a key. Once in a bus and once in a library, very annoying to others. I wish they would turn the feature off, but these non-geek users are probably thinking "it has to beep, it's electronic".
I certainly agree with this sentiment. For what it's worth, I think that it's a matter of finding the right balance. My Palm Pilot has a very unobtrusive click that I find very useful (for example, if I scroll accidentally); on the other hand, my cell phone has a louder beep that I often find distracting (however, I'm not often typing more than a few numbers on my cell so it's not as bad.)
Besides, if you've ever used a computer while listening to music with headphones, then you've been using a computer without hearings the clicks. Did you miss them? Audio feedback might not be as necessary as you think.
Actually, even though I almost always listen to music when on the computer, I almost never use headphones. (If I'm in a public place, I simply won't listen to headphones.) However, I'm not sure if this is actually relevant to the current discussion. Mostly, I think it's because I don't like being disconnected from my environment. In any case, it means that I haven't experienced what you're talking about.
The few times that I have used headphones while on the computer was for gaming (back when I used to have time to game). But in first person shooters, the game itself provides audio feedback so I never felt that I was missing anything.
Finally, I guess that I should say that I don't think that audio feedback is absolutely necessary. If that were the case, deaf people wouldn't be able to type. However, I do feel that multiple forms of feedback (when, as you point out, it is non-obtrusive) is advantageous.
Not to troll, but why do you want a no-click mouse in the first place? The clicking sound is audio feedback that improves usability. Same reason that keyboards click. More feedback (tactile, visual, audio) is more useful.
It's one thing if you're asking out of curiosity or just to see if you can do it. But if you think that there's a real, measurable benefit to a silent-click, I think that you might want to spend some time reconsidering.
I guess the general question would be: "Under what circumstances would having a silent-click be beneficial?"
I'm looking forward to it. I could never make any progress in Myst because I'm color-blind. Switching indicators from red to green? I don't think so. (Took me days to figure out what I was supposed to be trying to do.) Now I'll finally be able to figure out what the hell was going on!
Can some of the clever folks here at /. come up with a way for Open Source to succeed and pay programmers at the same time?
Nobody needs to come up with a "clever way;" it's already been done. The vast majority of software is produced under contract or for internal-use only.
This is often precisely the types of software that should be open-sourced: software that is responsible for a company's infrastructure but does not provide a competitive advantage. If a particular piece of software is that important to a company's business plan, I can understand why they would want to keep it to themselves. In most cases, however, the company would be better served by using (and contributing to) community-developed software - software that has been well tested, is extremely stable, and very inexpensive.
You don't need to produced shrinkwrapped software in order to make money. Rather, programmers should sell (rent) their services. In fact, to my mind, the idea that we wait for someone to produce the software that we want is ass-backwards. If there's a piece of software that I need, and I can't code it myself, I should just go hire someone to write it for me. (And, when I get back into grad school and have grant money available to me - that's precisely what I intend to do.)
I have the same problem. A few years ago, my [ex-]employer forced me to get an MSN IM account. Just recently, I've begun receiving unsolicited email from Microsoft at that address. What's particulary frustrating is that there is no way to unsubscribe from Passport!
Joel Spolsky argues that offices are essential for developers because it's the only way to get into "the zone." Without a door to close, cubicles invite interruption. Somebody innocently asks a question and bang! all of the variables, procedures, and methods drop out of your head. Takes a good 5, 10, 15 minutes or more to get back to find your place. Considering how much programmers are paid, that adds up to serious money.
0 68.html
Or, like I just did, you might just start reading Slashdot. I gotta get back to work...
http://www.joelonsoftware.com/articles/fog0000000
Actually, although his company, FogCreek Software, sells proprietary software, they do provide the customer with a copy of the source-code for some (all?) of their products. It's not Open Source (and certainly not Free Software) but you can review the code yourself, fix bugs, etc...
0 26.html
http://www.joelonsoftware.com/articles/fog0000000
It seems to be very much a YMMV type of thing. Also, it very much depends on what you're talking about. Mozilla's always been slow at starting up. (Preloading dlls does help things for Windows; also, under Windows, I install only the browser. I don't know if that matters or not.)
Regarding page rendering, I've found builds 0.9.8 and 0.9.9 to be incredibly fast. Even Slashdot! In fact, Mozilla renders pages significantly faster on my Pentium 200 Debian Potato system than IE does on my Pentium III 500 Win2K system. (I don't know if it's significant but for my Debian system, I use the prebuilt tar.gz package that doesn't include the Talkback app. Mozilla hasn't crashed on me in quite some time but I've had bad experiences with the Talkback builds in the past.)
I'm really not trying to troll here but anti-aliased text has always looked fuzzy to me. In the screen shots, for example, the spacing and sizing of the AA text is certainly nicer but the default text seems shaper and crisper. Am I wrong here? Joel Spolsky agrees with me but everyone else seems so excited about AA text that I have to wonder if I'm missing something.
I thought that this was great until I read the actual laws involved. The California Business & Professions Code 17538.45 (the law that the individual sued under) states that an Email Service Provider (ESP) may sue individuals and companies that send Unsolicited Commercial Email (UCE) over their servers.
The only reason that the individual was able to win her suit was because her email is forwarded to one machine to another which (in her words) 'may make [her] qualify as an electronic mail service provider'.
This is indeed a victory but it's a small one. The responsibility still lies with the ISPs. Most individuals won't qualify as an ESP and, therefore, can't sue.
The PCGuide at www.pcguide.com has a wonderful discussion of backup and disaster recovery procedures at http://www.pcguide.com/care/bu/index.htm. Discusses the various methods and mediums, rotation schedules, onsite vs offsite storage, etc. Also has an analysis of the how well various backup methods protect against various risks (i.e., hardware failure, file system corruption, accidental deletion). A must read.
As a graduate student in sociology, computing and programming is central to my work. This doesn't mean, however, that I practice computer-science. Computers are simply the tool which I used to conduct my work.
In my experience, most academics in the natural and social sciences are fairly good programmers. Among social scientists, we've all had to learn a variety of program-specific scripting languages (e.g. SPSS and SAS) in order to conduct our analyses. But few of us would count as (or consider ourselves to be) "hackers" or "compter scientists". And, while I've collected a number of programming languages over the years, they are simply the various tools that I use to collect and analyze data. Sed and awk are indispensible and Python has proved quite useful. I write all my papers in LaTeX and expect that I'll eventually learn enough Lisp to tame Emacs. But I'll probably never need C.
Even my father, who, as a graduate student at Stanford during the 60's, was writing Fortran programs in order to conduct his statistical analysis, has never considered himself a programmer. Fortran, too, was simply a tool.
We don't care -- or need to care -- about the theory underlying computing.
With all of the above being said, there is a current development within the social scientific academic community that cast further doubt upon the article's premise -- and one that I find rather worrisome. Rather than increasing, the programming abilities of the academic community (at least within the social sciences) may be declining. And, I believe, that this could have very serious consequences.
Being raised on GUIs, current graduate students are being taught to point-and-click their way through analysis. SPSS, for example, has, over the years, built up a rather nice nice GUI front-end. But, as in all GUIs, the SPSS front-end is less powerful and less flexible than the underlying scripting language.
But students aren't being taught the scripting languages. As a result, many students don't know what their programs are capable of or how to conduct complex analysis. I suspect that many students don't even realize that they can conduct more complex analysis! And, of course, most of the programs used to conduct complex analysis are entirely text based. What I find most worrisome, however, is that the GUIs serve to insulate the user from the analytical process -- as a result, students may be failing to understand and appreciate the statistical theory underlying their analysis.
As social scientists, we are not computer scientists; we don't need to understand computer theory. We do, however, need to understand statistical theory.
Actually, anyone who works professionally in Linux should be in favor of certifications. The entire point of certifications (and professional organizations) is to reduce the number of "qualified" applicants so that those who have the certifications can charge more for their services.
This isn't to say that certification procedures aren't crap and aren't meaningless -- but to focus upon that piece of the puzzle is to miss to point of what certifictions are really about.
Why do employees always (usually) believe that managers are assholes? For the same reason that managers always (usually) think that employees are lazy. It's inherent in the relationship.
The entire reason for management's existence is to increase the productivity of workers. Productivity can be increased in only two ways -- by inducing workers to work harder or by forcing workers to work harder.
We in the IT industry generally have it pretty good -- inducement is far more pleasurable than domination. But domination always comes into play at some point, somewhere. Furthermore, the threat of domination is omni-present. And that puts great strain on the relationship.