First of all, the actual number of Windows applications probably has little or nothing to do with the barriers to entry. A better metric might be file formats for different document types.
Second of all, whether there are 10, 100, 1000, or 1000000 applications, most people only use about six.
Slashdotters already know this. Else, why would Linux folks be bothering to try to duplication the "office suite" applications? Why? Because we know that once you get folks away from Word and Excel, the rest can follow.
Those two types of applications, plus a user interface that any random, semi-trained human can use would just about do it for the vast majority of Windows users.
Please think about that for a second: if the Linux world didn't agree with what I just said, then why are we bothering to build office suites or fix up the UI?
Back in my WABI days, we found that Delorme Street Atlas was so well-coded and clean that it ran with extraordinarily few problems on WABI.
I've never tried it on WINE, but I suspect that it would have a pretty good shot there, too, depending on the current status of the WINE implementation.
The reason we liked Delorme so much was that most of the applications we tried to run on WABI turned up at least a handful of compatibility issues that we had to deal with. Many MS applications took months of analysis (black box analysis, of course) to get them to work.
I spend quite a bit of my time interviewing and hiring engineers, web designers, writers, and other folks as well.
I have never asked a candidate to sign an NDA, and I usually try to give them just enough information to evaluate the job without learning too much.
However, it's important to remember that an interviewer has basically three jobs to do when they interview you:
Assess whether you are technically able to do the work.
Assess whether your personality will fit in with the other members of the team.
Sell you on the company and the project, so that you want to work there if they decide to give you an offer.
It can be extraordinarily hard, sometimes, to sell you on how exciting the project will be without telling you some things that you "don't need to know." Often, the more gee-whiz the project, the more difficult it is to give a true representation of what you'll be doing without the NDA.
Should you sign one? Well, that's for you to figure out for yourself, and it probably depends on the project and how much inherent trust you have for the employer. If at this early stage you don't feel enough trust for them to sign an NDA, it may be a sign that they're really not the best match for you.
But remember that although there are definitely employers that throw non-disclosure agreements around as though they had found the location of the holy ark of the covenant, there are always a few who legitimately have something novel and what to share it with you in a safe setting.
This project sound remarkably like WABI and WINE, two projects (one commercial, one open source) that were meant to duplicate the Win APIs well enough to allow applications written for them to run on other systems.
This was attempted back in the late 80's and early 90's by several companies, including a few that I didn't mention above. (I was on the WABI project and most of the founders of the company who started it are still working for Sun.)
This is not the same, unfortunately, as actually porting the applications. Because the Win APIs cover everything from painting the screen to dealing with devices, this kind of work is much more like emulating an O/S than like true porting.
Also, if you've seen how Win applications deal with their environment (most of them were very poor neighbors, system-wise), you would probably agree that since they were never built to be ported to other O/S's, most of them would be a real bear to try to move.
I don't want to be pessimistic, but until and unless Microsoft assigns its own army of coders (instead of just licensing source to other people), I give this project relatively low odds for success.
From Microsoft's side, it's a no-lose. Without substantially damaging their core O/S, they get a shot at having someone else do the heavy lifting, plus -- because they'd be under license agreement to MS -- MS could always yank back on them if things get "out of hand." In the likely case that nothing comes of it, MS at least gets to pocket some licensing fees. A reasonably clever move for them, but one I'm sure won't be popular on/.
The reason that a book like Cyberselfish gets published and reviewed and debated (to death, apparently) is that Libertarian thought is no longer fringe, no longer unusual, and no longer running against the tide of history.
On the contrary, Libertarian thought and the movement away from massive government control of economies and personal choice is extraordinarily relevant.
Were the old guard not frightened of it, they'd see no reason to complain about it.
Dogs don't bark at tombstones, and publishers don't publish books on irrelevant movements.
...I just can't believe that people are smart enough to make big decisions on their own...
Then how can you trust other people to make the decisions for them? Because that's all that big government is: a collection of other people to make decisions for you (and then enforce them).
Admittedly, we're all merely human. But if you can't make decisions for your own life, then how has anyone else even got a prayer of getting it right for you?
Of course, the government is needed for some things (national defense and the court system being only two obvious examples), but basing your system of government on the idea that people are imbeciles doesn't seem to me to be viable in the long run.
Haven't we already beat the stuffing out of Ms. Barsook's rambling, understanding-free diatribe?
This stuff is old news, though it is reliable flame-bait. Perhaps/. needs more hits this week.
Re:These kids never saw a recession
on
Selfish Society
·
· Score: 1
For these reasons, this year I'm voting Ralph Nader for President.
Let me get this straight: You're worried about recession, so you're going to vote for the candidate from the most interventionist and anti-marketplace party on the list?
Talk about self-fulfilling prophecies!
Boy, they'd sure better talk about safety nets. The minute they started in on their policies, we'd all need one.
Re:How can anyone call us selfish?
on
Selfish Society
·
· Score: 1
How can anyone call us selfish?
Easy. Because "selfish" is a code word. It's meaning, in this context, goes something like this:
selfish (sel'fish), adj, 1. a person who stubbornly refuses to do what I want them to, regardless of how good it would be for me. 2. characterized by or revealing a desire to keep the money one has earned honestly, make one's own decisions, and take responsibility for one's own life. 3. opposed to big government and/or big corporations. 4.Libertarian.
What a shame that this article starts out with such promise and then fails so miserably.
Of course, it's silly to think that software shouldn't be sold just because the marginal cost of making the second and succeeding copies is very low. Happily, that idea survives only in the writings of RMS, and is generally ignored by the rest of the software community (including most of the folks who are involved in the open source community, who also have day-jobs, for the most part).
What's most unforgivable about this article, however, is the way that it takes RMS to task for behaving publicly and privately as if software sellers are evil incarnate, and for general rudeness in the face of those who disagree, while simultaneously making ridiculous and rude ad hominem attacks on ESR because of his beliefs about gun rights. This article's vituperative and wholly uncalled for moralizing on the topic of firearms is both childish and off-topic.
(Even more interesting, had the author investigated *why* ESR believes in gun rights, he would have learned that ESR is a Libertarian and as such would not agree that software has to be given away for free. But why let the facts get in the way of a good argument?)
In the end, time has burned away the excesses of the "free software" movement and left in its place a tremendous achievement: proof positive that you can make money in software without necessarily keeping the code a trade secret.
Not only is this a boon to software developers, it also enables open source projects to ramp up and evolve more quickly than many of their closed counterparts.
Had the author focused on these aspects, rather than getting exercised about RMS's behavior at a dinner party, or the fact that ESR's gun talk gives him the willies, this might have been a much better article.
Slashdot Mirror
Second of all, whether there are 10, 100, 1000, or 1000000 applications, most people only use about six.
Slashdotters already know this. Else, why would Linux folks be bothering to try to duplication the "office suite" applications? Why? Because we know that once you get folks away from Word and Excel, the rest can follow.
Those two types of applications, plus a user interface that any random, semi-trained human can use would just about do it for the vast majority of Windows users.
Please think about that for a second: if the Linux world didn't agree with what I just said, then why are we bothering to build office suites or fix up the UI?
Back in my WABI days, we found that Delorme Street Atlas was so well-coded and clean that it ran with extraordinarily few problems on WABI.
I've never tried it on WINE, but I suspect that it would have a pretty good shot there, too, depending on the current status of the WINE implementation.
The reason we liked Delorme so much was that most of the applications we tried to run on WABI turned up at least a handful of compatibility issues that we had to deal with. Many MS applications took months of analysis (black box analysis, of course) to get them to work.
Delorme Street Atlas worked right out of the box.
Of course, YMMV, this was 5+ years ago.
I have never asked a candidate to sign an NDA, and I usually try to give them just enough information to evaluate the job without learning too much.
However, it's important to remember that an interviewer has basically three jobs to do when they interview you:
It can be extraordinarily hard, sometimes, to sell you on how exciting the project will be without telling you some things that you "don't need to know." Often, the more gee-whiz the project, the more difficult it is to give a true representation of what you'll be doing without the NDA.
Should you sign one? Well, that's for you to figure out for yourself, and it probably depends on the project and how much inherent trust you have for the employer. If at this early stage you don't feel enough trust for them to sign an NDA, it may be a sign that they're really not the best match for you.
But remember that although there are definitely employers that throw non-disclosure agreements around as though they had found the location of the holy ark of the covenant, there are always a few who legitimately have something novel and what to share it with you in a safe setting.
This project sound remarkably like WABI and WINE, two projects (one commercial, one open source) that were meant to duplicate the Win APIs well enough to allow applications written for them to run on other systems.
This was attempted back in the late 80's and early 90's by several companies, including a few that I didn't mention above. (I was on the WABI project and most of the founders of the company who started it are still working for Sun.)
This is not the same, unfortunately, as actually porting the applications. Because the Win APIs cover everything from painting the screen to dealing with devices, this kind of work is much more like emulating an O/S than like true porting.
Also, if you've seen how Win applications deal with their environment (most of them were very poor neighbors, system-wise), you would probably agree that since they were never built to be ported to other O/S's, most of them would be a real bear to try to move.
I don't want to be pessimistic, but until and unless Microsoft assigns its own army of coders (instead of just licensing source to other people), I give this project relatively low odds for success.
From Microsoft's side, it's a no-lose. Without substantially damaging their core O/S, they get a shot at having someone else do the heavy lifting, plus -- because they'd be under license agreement to MS -- MS could always yank back on them if things get "out of hand." In the likely case that nothing comes of it, MS at least gets to pocket some licensing fees. A reasonably clever move for them, but one I'm sure won't be popular on /.
Check out my current client: hireability.com
I agree that this is not the only way (or even the best way) to choose a political affiliation.
However, if you want to know which party "gets" the Open Source movement, there's really only one answer.
On the contrary, Libertarian thought and the movement away from massive government control of economies and personal choice is extraordinarily relevant.
Were the old guard not frightened of it, they'd see no reason to complain about it.
Dogs don't bark at tombstones, and publishers don't publish books on irrelevant movements.
Then how can you trust other people to make the decisions for them? Because that's all that big government is: a collection of other people to make decisions for you (and then enforce them).
Admittedly, we're all merely human. But if you can't make decisions for your own life, then how has anyone else even got a prayer of getting it right for you?
Of course, the government is needed for some things (national defense and the court system being only two obvious examples), but basing your system of government on the idea that people are imbeciles doesn't seem to me to be viable in the long run.
This stuff is old news, though it is reliable flame-bait. Perhaps /. needs more hits this week.
Let me get this straight: You're worried about recession, so you're going to vote for the candidate from the most interventionist and anti-marketplace party on the list?
Talk about self-fulfilling prophecies!
Boy, they'd sure better talk about safety nets. The minute they started in on their policies, we'd all need one.
Easy. Because "selfish" is a code word. It's meaning, in this context, goes something like this:
What a shame that this article starts out with such promise and then fails so miserably.
Of course, it's silly to think that software shouldn't be sold just because the marginal cost of making the second and succeeding copies is very low. Happily, that idea survives only in the writings of RMS, and is generally ignored by the rest of the software community (including most of the folks who are involved in the open source community, who also have day-jobs, for the most part).
What's most unforgivable about this article, however, is the way that it takes RMS to task for behaving publicly and privately as if software sellers are evil incarnate, and for general rudeness in the face of those who disagree, while simultaneously making ridiculous and rude ad hominem attacks on ESR because of his beliefs about gun rights. This article's vituperative and wholly uncalled for moralizing on the topic of firearms is both childish and off-topic.
(Even more interesting, had the author investigated *why* ESR believes in gun rights, he would have learned that ESR is a Libertarian and as such would not agree that software has to be given away for free. But why let the facts get in the way of a good argument?)
In the end, time has burned away the excesses of the "free software" movement and left in its place a tremendous achievement: proof positive that you can make money in software without necessarily keeping the code a trade secret.
Not only is this a boon to software developers, it also enables open source projects to ramp up and evolve more quickly than many of their closed counterparts.
Had the author focused on these aspects, rather than getting exercised about RMS's behavior at a dinner party, or the fact that ESR's gun talk gives him the willies, this might have been a much better article.