Isn't this stuff supposed to get _better_ with time? What ever happened to "release early, release often"?
Does Microsoft "release early, release often"? Can you think of any major software project that does? As a developer, I like the sentiment, and we try to follow it in our own projects, but in large, complex systems like Gnome, it's unrealistic to expect an unending frequent stream of updates. Similarly, installing all the latest updates as soon as they're released is a recipe for frustration, again with any software.
If you want stability, pick a stable version and stay with it. If you want the excitement of continally experimenting with the latest stuff, then you have to live with bugs and instability. A "stable" release just means that it's been tested by beta testers, it's not truly stable until it's been tested by everyone else, too, and then patched with the necessary fixes.
It's your choice, don't make it and then complain about it once you experience the consequences.
> I know that as SOON as I can, I'm going to acquire
> expertise in a UNIX other than Solaris or Linux,
> and in a context other than the Internet.
Quite right. Solaris was used in the dot-com boom because VCs felt that it covered their asses. It had nothing to do with technical strength and everything to do with "Nobody ever got reamed out by their investors for using Solaris". When you hear how it cost "donthaveabusinessmodel.com" $10m to set up their web servers and store application, that's because the VCs and their lackeys spec'd Solaris, Oracle, and some nice expensive old-school brand-name non-commodity Unix hardware. Woulda cost $25K max if it had been done in any non-dot-com business.
In the real world, there are many businesses that turn over billions of dollars annually, have heavy-duty information processing requirements, yet run entirely on Windows LANs, for example. These aren't Fortune 500 companies, they're just medium-sized. The problem is that dot-coms tried to emulate the Amazon "instant-Fortune-500" model, which ended up with the ridiculous scenario of little companies with 50 people acting as though they were General Electric. "Do you think we need an HP Superdome, or will a farm of Sun Starfires do the trick?" Business sense didn't go out the window - it was never there to begin with.
My advice to people in this market is to learn a bit about business yourself (read the Wall Street Journal or The Economist or something) and make your own, realistic assessment of the business that's hiring you. How do they make their money? Who are their customers? How are they going to survive the next economic downturn? Also, try to determine what value you're adding to the business. Being a kick-ass sysadmin means nothing if you're administering machines that aren't necessary in the first place.
Many people tend to accept jobs with a charming but foolish degree of trust. They trust that the company really needs them and that they're not being hired due to some political whim. They trust that the company isn't going to go out of business within the next 12 months ("Surely they would warn me?") Trusting your employer to know what's best for you, or even for themselves, is a 1950's hangover from when IBM employees all got together every morning in their suits with mandatory white shirts to sing inspirational company songs. Get with it, people! You're an independent economic entity and the only person responsible for your economic survival is YOU! If you take a job at a doomed company, you have nobody to blame but yourself!
I applaud the "no overkill" philosophy and am encouraged to hear of sensible, successful companies in this space.
Now in the interests of "no overkill", could you get the company's web designers to change the font size on the home page to something in points, not pixels? The current "12px" setting comes out at about a 6 point size on many high-resolution, high-pixel-density screens. Print it out on a laser printer and see what it looks like... Now where did I put that magnifying glass???
>...who would volunteer to run a forum at MSN!!!!!
Probably the people you pissed off with your elitist attitude...
Learn Scheme & Computer Science at the same time
on
Learn A New Language
·
· Score: 2
Scheme is a very elegant member of the LISP family that's used to teach and explore advanced programming and computer science at many universities, including MIT, Yale etc. The basics of Scheme are easy to learn, and the language is mind-blowingly powerful. Learning Scheme is guaranteed to expand your programming horizons. Plus, there's a wealth of good Scheme tutorial information on the web:
The Scheme Language Standard, "R5RS", is actually surprisingly readable, at about 50 pages. Here's a
PDF of R5RS, which is useful to have around if you're actually using the language.
The Structure and Interpretation of Computer Programs (SICP) - a.k.a. the wizard book, a very worthwhile and enlightening book, used as an introductory text at MIT. It's available on the web, but it's not a light or quick read.
How to Design Programs (HTDP) - a book that teach program design techniques, using Scheme.
Learning Scheme will teach you many programming language concepts which will help you understand other languages - even the ones you already know - better. The only downside is that you'll start to wish that other languages had some of Scheme's more powerful features. Perl, Python and Javascript have all adopted some Scheme-like features in the last several years, and after learning Scheme, you'll understand these features better (e.g. closures, continuations, first-class functions).
Listen, if you're really paranoid about security, or even just responsible, you would not be running a public web server on IIS5 at this point in time.
Even IIS4 is only just getting to the point where it can be considered reasonably secure, although you have to apply various security patches over and above the last service pack to get it to that point.
I can just hear the wannabes saying "but IIS5 is what comes on Win2K! We can't run IIS5!" Simple answer to that: don't run Win2K on public web servers, either! If you have to run a closed source, Microsoft web server, run IIS4 on NT4. After IIS5 has been out for another year or so, once most of the security violations have been found and patched, then you can start considering migrating to it.
If you're not that patient, switch to a nice stable version of Apache. You know you want to.
Magellan? What kind of nerd/geek are you?! That's for consumers, dammit!
Get yourself a PDA - e.g. Pilot or Cassiopeia (color display is nice for maps) - along with a handlebar mount (not sure how well that'll work on the BikeE, I have a Vision R-45 "real" recumbent myself.) Hook the PDA up to a GPS unit - there are a few of them that will do the trick, including Magellan - any one with an NMEA interface, which is most of them, is theoretically hook-uppable, but you don't need one with any fancy user interface features. Then use mapping software like Street Atlas USA (Solus is the name of their PDA app, which you can buy in a bundle with a GPS). Now you have something much more flexible, you can dock with your PC to set up routes instead of dealing with a brain-dead consumer GPS unit, you can easily upload the records of where you went into Street Atlas on your PC, etc., and in general the geek factor will be an order of magnitude higher!
I think the problem in Connecticut is that it's a state chock-full of high-powered executives who are too busy making billions of dollars on a global scale (financial, insurance and tech firms, among others) to worry about petty matters like local state government. Hence, state government is run by people who have nothing better to do. Ms. Harp seems like a typical case in point.
Is it just me, or does this author seem incredibly petty? Whether Cliff did or did not "get" his book, this level of response to a review exhibits a staggering degree of insecurity on the part of the author.
He might have posted a message which explained his book in a more positive light, rather than picking away at every comment Cliff made.
I read the earlier review and wasn't really swayed one way or the other as to whether the book should interest me - it certainly didn't grab me. Having read this rebuttal, I know for sure I don't want to waste any time on this guy's stuff.
Hey, if y'all had lived through a four-hour power failure last night, like I did, you'd think this was funny too! I'm off to shop for a fuel cell...
Re:Problem with the FAQ...
on
GPL FAQ
·
· Score: 1
Why can't the author of GPL'd software assign power of attorney (or something similar) over their copyright to the FSF? That would seem to better retain the spirit of credit where credit is due, not to mention allow the author to retain control over use of her code for other purposes.
The "here is that 411" message in this thread describes this - the project given away by a Russian double agent related to an inductive tap of an old-fashioned undersea copper cable in the Okhotsk Sea.
Not even the NSA can tap fiberoptics inductively - laws of physics and all that. They would have to splice it, a much more difficult thing to do at the bottom of the ocean.
If a fiberoptic tap has really occurred - and as far as I can tell, the evidence is simply from unnamed sources according to ZDNet - it would be a very different animal from the Okhotsk tap. Okhotsk used high-capacity recorders to store the info for later retrieval by submarine. That would have been analog data. You couldn't save enough digital fiberoptic data in a recording pod to make it worthwhile. You'd have to drop a Cray on the seafloor to process some of the data in realtime and save only what you're interested in.
That's an operation for a l33t hax0r somewhere - hack into the NSA Cray that's sitting on the ocean bed somewhere off the Kamchatka pensinsula...
btw, the reason cars that go fast are impressive is because people don't want to drive a motorcycle. A car is a hell of a lot more comfortable, a hell of a lot more safer, much more immune to driving under adverse conditions, and far more simpler to operate safely than a bike.
Huh? Motorcycles are for *fun*. Sure, some kids drive them as their only vehicles, like I did when I was at university. But saying a car is a whole lot safer than a motorcycle is like saying an ocean liner is a whole lot safer than a scuba tank. Doesn't help you if what you're interested in is a close-up view of the humuhumunukunukuapua'a (Hawaii's state fish).
My original reply was to a message that was comparing the performance a motorcycle to a car; I was pointing out that for the purposes of performance comparisons, you have to compare bikes to bikes. Use of the term "lame" was what is known as a rhetorical device (adjusted for/.!) Your prejudice against motorcycles has nothing to do with this discussion. Many sports are dangerous, especially if done irresponsibly, including skiing, climbing, diving, and biking (with or without an engine). If you want total safety, stay in bed and make sure you don't live near a fault line. Oh, and cut back on those fatty foods.
There are faster bikes out there, especially if you go custom/turbocharged etc. So this rocket bike is fast, but not *that* fast. And if you've ever done more than about 140mph on a bike, you know that a top end of 250mph is kinda academic.
Yeah, but as in my original example, if the product you're looking for is a specific brandname item, as opposed to a commodity product, the sources for that product may be limited to a single manufacturer and its dealers, so "seeking out a better price" may be easier said than done.
Your argument sounds like one of "buyer beware", which is always good advice, but I'm talking about customer relations, and saying that vendors who ignore these issues are being short-sighted.
My point is that if companies are going to be changing prices from moment to moment, they should help the customer by giving some insight into what's going on. Having to go to a third-party to figure out things that your vendor isn't telling you isn't conducive to good customer relations.
Nice Zen-like question, but the answer is easier than the one about a tree falling in a forest: if the discrimination can be kept secret from all customers forever, the company can "get away with it", whether or not you consider harm to have been done.
But in real life, things like this can do enormous harm because it isn't really possible to keep it secret. If you found out that someone else had bought the same product that you did from the same place on the same day at a lower price, how would you feel about it?
You're right, this is normal. However, in the offline world, it's not uncommon for a salesman to tell you that a price is about to or has just gone up or down. On the web, it's very unusual to see any notifications about such things, except in the case of specially discounted items or sales.
What would be good from the customer perspective is if websites actually provided some pricing rationale and history. You could click on an item's price and see that yes, it is $100 more expensive today than yesterday, but that's because the price of RAM has just gone up, say.
Otherwise, pricing is just a black box and customers have no way of knowing if they're being discriminated against.
Of course, there are tools to help customers compare prices across web sites, so in an absolute sense, it's not a problem. But vendors would be wise to consider the impression that these things leave on customers. If I want a Thinkpad specifically, I can't go to anyone but IBM or an IBM dealer to get it. If I suspect IBM is playing funky games with pricing, I may decide I'm better off with someone else.
In short, transparency is a good thing, in pricing as in many other areas.
I don't see anything about this on the home page of the EFF's web site. The Slashdot submission isn't clear about what they're asking for and why, although they imply it's being gathered "just in case". They don't even define what they mean by P2P, and that matters. After all, plenty of people have websites on their home machines these days, hooked up 24x7 via cable or DSL. These websites can be indexed by search engines. That's P2P. It's also a kind of P2P that the government can't easily screw with - you'd have to disable the search engines, or do something truly idiotic like outlaw private web servers.
I'm guessing this submission was just a moronic employee at EFF who overheard something in the hallways and decided "hey - I bet/. would be great for this!"
To use the words of the submission, "demerits applied" for most useless information request from an organization that should know better.
...is the Net itself, in cases like this. Wherever governments try to police their citizen's thoughts, the Net will be a subversive force. Read Shockwave Rider one more time and rest easy in the knowledge that Mumbai's citizens will soon be free again.
Don't you see? It's a coverup! Doug Adams isn't dead - he obviously had to leave Earth because it is about to be blown up (by the Vogons, who else?) to make way for an intergalactic information superhighway. Probably the only reason we're still alive right now is that the Vogons are busy arguing on the bridge of the spaceship about whether to destroy the entire solar system at once, or one planet at a time.
And if the coverup isn't enough proof for you, the fact that the dolphins are leaving should clinch it.
Leave now, while you still can!
Spork, why are you so threatened by Open Source?
on
The Secret of Life
·
· Score: 1
Don't worry, we'll make space for you to do something mindless and menial in this brave new open source world we're building. "Give up us your trolls, your sporks, your huddled masses!"
An inner class might constitute a closure, but to use that to program in functional style is a PITA - I know, I'm currently busy converting a C++ program that relies heavily on functors to parameterize its behavior, into Haskell. I wrote the C++ code some years back, and while it was fast and did the job, maintaining and extending it was a pain.
In Java or C++, classes (not just inner classes) can be used to emulate the behavior of both closures and first-class procedures in order to code in a somewhat functional style. But that doesn't qualify a language as having functional features. My point was that in Smalltalk, closures and first-class procedures were already watered down, and Java all but eliminated them. Perl and Javascript both have real first-class procedures and real closures, and they are both stronger languages for that.
I agree that a hybrid language isn't ideal, but I was predicting what I think will happen, not what I'd like to see happen. I don't yet see a functional language that's truly ready to completely replace imperative languages for average programmers. So for the forseeable future, the mainstream languages will simply adopt functional features.
Slashdot Mirror
Does Microsoft "release early, release often"? Can you think of any major software project that does? As a developer, I like the sentiment, and we try to follow it in our own projects, but in large, complex systems like Gnome, it's unrealistic to expect an unending frequent stream of updates. Similarly, installing all the latest updates as soon as they're released is a recipe for frustration, again with any software.
If you want stability, pick a stable version and stay with it. If you want the excitement of continally experimenting with the latest stuff, then you have to live with bugs and instability. A "stable" release just means that it's been tested by beta testers, it's not truly stable until it's been tested by everyone else, too, and then patched with the necessary fixes.
It's your choice, don't make it and then complain about it once you experience the consequences.
> expertise in a UNIX other than Solaris or Linux,
> and in a context other than the Internet.
Quite right. Solaris was used in the dot-com boom because VCs felt that it covered their asses. It had nothing to do with technical strength and everything to do with "Nobody ever got reamed out by their investors for using Solaris". When you hear how it cost "donthaveabusinessmodel.com" $10m to set up their web servers and store application, that's because the VCs and their lackeys spec'd Solaris, Oracle, and some nice expensive old-school brand-name non-commodity Unix hardware. Woulda cost $25K max if it had been done in any non-dot-com business.
In the real world, there are many businesses that turn over billions of dollars annually, have heavy-duty information processing requirements, yet run entirely on Windows LANs, for example. These aren't Fortune 500 companies, they're just medium-sized. The problem is that dot-coms tried to emulate the Amazon "instant-Fortune-500" model, which ended up with the ridiculous scenario of little companies with 50 people acting as though they were General Electric. "Do you think we need an HP Superdome, or will a farm of Sun Starfires do the trick?" Business sense didn't go out the window - it was never there to begin with.
My advice to people in this market is to learn a bit about business yourself (read the Wall Street Journal or The Economist or something) and make your own, realistic assessment of the business that's hiring you. How do they make their money? Who are their customers? How are they going to survive the next economic downturn? Also, try to determine what value you're adding to the business. Being a kick-ass sysadmin means nothing if you're administering machines that aren't necessary in the first place.
Many people tend to accept jobs with a charming but foolish degree of trust. They trust that the company really needs them and that they're not being hired due to some political whim. They trust that the company isn't going to go out of business within the next 12 months ("Surely they would warn me?") Trusting your employer to know what's best for you, or even for themselves, is a 1950's hangover from when IBM employees all got together every morning in their suits with mandatory white shirts to sing inspirational company songs. Get with it, people! You're an independent economic entity and the only person responsible for your economic survival is YOU! If you take a job at a doomed company, you have nobody to blame but yourself!
Now in the interests of "no overkill", could you get the company's web designers to change the font size on the home page to something in points, not pixels? The current "12px" setting comes out at about a 6 point size on many high-resolution, high-pixel-density screens. Print it out on a laser printer and see what it looks like... Now where did I put that magnifying glass???
Just a pet peeve of mine...
Probably the people you pissed off with your elitist attitude...
- "Teach Yourself Scheme in Fixnum Days": A comprehensive tutorial
- "The Scheme Language": a shorter tutorial
- "Invitation to Scheme": yet another tutorial
- The Scheme Language Standard, "R5RS", is actually surprisingly readable, at about 50 pages. Here's a
PDF of R5RS, which is useful to have around if you're actually using the language.
- The Structure and Interpretation of Computer Programs (SICP) - a.k.a. the wizard book, a very worthwhile and enlightening book, used as an introductory text at MIT. It's available on the web, but it's not a light or quick read.
- How to Design Programs (HTDP) - a book that teach program design techniques, using Scheme.
Learning Scheme will teach you many programming language concepts which will help you understand other languages - even the ones you already know - better. The only downside is that you'll start to wish that other languages had some of Scheme's more powerful features. Perl, Python and Javascript have all adopted some Scheme-like features in the last several years, and after learning Scheme, you'll understand these features better (e.g. closures, continuations, first-class functions).Even IIS4 is only just getting to the point where it can be considered reasonably secure, although you have to apply various security patches over and above the last service pack to get it to that point.
I can just hear the wannabes saying "but IIS5 is what comes on Win2K! We can't run IIS5!" Simple answer to that: don't run Win2K on public web servers, either! If you have to run a closed source, Microsoft web server, run IIS4 on NT4. After IIS5 has been out for another year or so, once most of the security violations have been found and patched, then you can start considering migrating to it.
If you're not that patient, switch to a nice stable version of Apache. You know you want to.
Get yourself a PDA - e.g. Pilot or Cassiopeia (color display is nice for maps) - along with a handlebar mount (not sure how well that'll work on the BikeE, I have a Vision R-45 "real" recumbent myself.) Hook the PDA up to a GPS unit - there are a few of them that will do the trick, including Magellan - any one with an NMEA interface, which is most of them, is theoretically hook-uppable, but you don't need one with any fancy user interface features. Then use mapping software like Street Atlas USA (Solus is the name of their PDA app, which you can buy in a bundle with a GPS). Now you have something much more flexible, you can dock with your PC to set up routes instead of dealing with a brain-dead consumer GPS unit, you can easily upload the records of where you went into Street Atlas on your PC, etc., and in general the geek factor will be an order of magnitude higher!
I think the problem in Connecticut is that it's a state chock-full of high-powered executives who are too busy making billions of dollars on a global scale (financial, insurance and tech firms, among others) to worry about petty matters like local state government. Hence, state government is run by people who have nothing better to do. Ms. Harp seems like a typical case in point.
He might have posted a message which explained his book in a more positive light, rather than picking away at every comment Cliff made.
I read the earlier review and wasn't really swayed one way or the other as to whether the book should interest me - it certainly didn't grab me. Having read this rebuttal, I know for sure I don't want to waste any time on this guy's stuff.
Hey, if y'all had lived through a four-hour power failure last night, like I did, you'd think this was funny too! I'm off to shop for a fuel cell...
Why can't the author of GPL'd software assign power of attorney (or something similar) over their copyright to the FSF? That would seem to better retain the spirit of credit where credit is due, not to mention allow the author to retain control over use of her code for other purposes.
All the NSA would have to do would be arrange for a fishing boat to snag the cable they were planning to splice, to explain the interruption.
Not even the NSA can tap fiberoptics inductively - laws of physics and all that. They would have to splice it, a much more difficult thing to do at the bottom of the ocean.
If a fiberoptic tap has really occurred - and as far as I can tell, the evidence is simply from unnamed sources according to ZDNet - it would be a very different animal from the Okhotsk tap. Okhotsk used high-capacity recorders to store the info for later retrieval by submarine. That would have been analog data. You couldn't save enough digital fiberoptic data in a recording pod to make it worthwhile. You'd have to drop a Cray on the seafloor to process some of the data in realtime and save only what you're interested in.
That's an operation for a l33t hax0r somewhere - hack into the NSA Cray that's sitting on the ocean bed somewhere off the Kamchatka pensinsula...
See subject.
Huh? Motorcycles are for *fun*. Sure, some kids drive them as their only vehicles, like I did when I was at university. But saying a car is a whole lot safer than a motorcycle is like saying an ocean liner is a whole lot safer than a scuba tank. Doesn't help you if what you're interested in is a close-up view of the humuhumunukunukuapua'a (Hawaii's state fish).
My original reply was to a message that was comparing the performance a motorcycle to a car; I was pointing out that for the purposes of performance comparisons, you have to compare bikes to bikes. Use of the term "lame" was what is known as a rhetorical device (adjusted for /.!) Your prejudice against motorcycles has nothing to do with this discussion. Many sports are dangerous, especially if done irresponsibly, including skiing, climbing, diving, and biking (with or without an engine). If you want total safety, stay in bed and make sure you don't live near a fault line. Oh, and cut back on those fatty foods.
The Yamaha Yzf-ri will do 10.15 seconds.
There are faster bikes out there, especially if you go custom/turbocharged etc. So this rocket bike is fast, but not *that* fast. And if you've ever done more than about 140mph on a bike, you know that a top end of 250mph is kinda academic.
Your argument sounds like one of "buyer beware", which is always good advice, but I'm talking about customer relations, and saying that vendors who ignore these issues are being short-sighted.
My point is that if companies are going to be changing prices from moment to moment, they should help the customer by giving some insight into what's going on. Having to go to a third-party to figure out things that your vendor isn't telling you isn't conducive to good customer relations.
But in real life, things like this can do enormous harm because it isn't really possible to keep it secret. If you found out that someone else had bought the same product that you did from the same place on the same day at a lower price, how would you feel about it?
What would be good from the customer perspective is if websites actually provided some pricing rationale and history. You could click on an item's price and see that yes, it is $100 more expensive today than yesterday, but that's because the price of RAM has just gone up, say.
Otherwise, pricing is just a black box and customers have no way of knowing if they're being discriminated against.
Of course, there are tools to help customers compare prices across web sites, so in an absolute sense, it's not a problem. But vendors would be wise to consider the impression that these things leave on customers. If I want a Thinkpad specifically, I can't go to anyone but IBM or an IBM dealer to get it. If I suspect IBM is playing funky games with pricing, I may decide I'm better off with someone else.
In short, transparency is a good thing, in pricing as in many other areas.
I'm guessing this submission was just a moronic employee at EFF who overheard something in the hallways and decided "hey - I bet /. would be great for this!"
To use the words of the submission, "demerits applied" for most useless information request from an organization that should know better.
...is the Net itself, in cases like this. Wherever governments try to police their citizen's thoughts, the Net will be a subversive force. Read Shockwave Rider one more time and rest easy in the knowledge that Mumbai's citizens will soon be free again.
And if the coverup isn't enough proof for you, the fact that the dolphins are leaving should clinch it.
Leave now, while you still can!
Don't worry, we'll make space for you to do something mindless and menial in this brave new open source world we're building. "Give up us your trolls, your sporks, your huddled masses!"
In Java or C++, classes (not just inner classes) can be used to emulate the behavior of both closures and first-class procedures in order to code in a somewhat functional style. But that doesn't qualify a language as having functional features. My point was that in Smalltalk, closures and first-class procedures were already watered down, and Java all but eliminated them. Perl and Javascript both have real first-class procedures and real closures, and they are both stronger languages for that.
I agree that a hybrid language isn't ideal, but I was predicting what I think will happen, not what I'd like to see happen. I don't yet see a functional language that's truly ready to completely replace imperative languages for average programmers. So for the forseeable future, the mainstream languages will simply adopt functional features.