Using IPv6 won't change that. It would technically mean we have an abundance of addresses, but our ISPs would still pull the same BS, expecting us to pay more for the same level of service.
Possibly not.
Back in Ye Olden Days, IP addresses were free and easy to get. But they became a relatively scarce resource, and companies started charging because of that.
When IPv6 takes off, ISPs will be able to give out as many addresses as they like without incurring significant costs. With even a modicum of competition, that will stop them from charging for extra addresses. And it certainly means they won't have to charge you for a single real address. If nothing else, that will result in reduced support calls.
Mind you, that may not apply to fixed addresses. People who need fixed addresses are up to something more than the average consumer. From the business perspective, it makes sense to charge more to them.
No. LEC. Local Exchange Carrier. It's in the FCC tariffs and all.
Sorry, the people I know talk about them collectively as LECs and separately as the ILEC (for incumbent) vs the CLECs. For an example of this usage, see this interesting prediction.
And yes, we pay $1700/month for a local DS3 on top of those fees to carry the circuits to us from the telco.
Wow. That's just robbery. $37.50 per month just to use the pair plus a slice of the DSLAM? Especially given that they also charge a monthly fee for the phone line and would have to maintain the wires anyhow? Ridiculous.
Again, here I am, the sysadmin for SCO. [...] I'm not the devil on Darl's shoulder telling him "Psst, ok, now sue these guys!". [...] would you still pass me by for a less qualified applicant.
Probably. I can teach skills. Fixing your ethical handicaps is beyond me.
SCO is a good example. Spammers are another one. I would expect people who work for them to be ethically challenged. Either that or unaware of what their employers were up to, in which case they'd be too clueless to bother with.
Why again, would hiring a kickass forward thinking ex-SCO sysadmin [...] not be in the "best interests of their firms"
The overly dramatic choice you set up, between the inadequate but nice employee or the skilled former concentration camp guard, isn't the one hiring managers face. The reality is that the manager will end up with a bunch of people who will probably be fine; the challenge is in picking the best one in the long term.
There are a few reasons I'd lean away from somebody with an ethically suspect past.
One would be that ethical problems are hard to detect. If somebody is incompetent, I'll know in pretty short order. If somebody is casing the joint for embezzlement opportunities, I might not know until the bank account is empty. Or, less dramatically, I might not realize that he's really clever about covering up shoddy work.
The second, and probably the biggest for me, would be a concern that the person won't get the big picture. A person who doesn't mind making a living spamming can't have much sympathy for the fucking colossal amount of trouble they are causing for their recipients. Why would that person care any more about the end users of the new company's products?
And the third would be simple CYA. Even if there's a relatively small chance that a person working for a corrupt company is themselves corrupt, what happens if it turns out to be true? Then suddenly not only am I the guy who hired the thief, but I'm the guy who hired the thief who used to the sysadmin for Enron's accounting department. Hindsight will make me look like a chump.
One of the least talked about reasons that everything has fallen apart in the tech sector is the sheer worthlessness of so many of the people in it.
This is very true, and it doesn't apply just to the line programmers.
A gold rush mentality meant many people made a lot of decisions that worked in the short term but screwed them in the long term. The theory at the time was that if things worked you'd get VC money (or go IPO) and have so much dough that you could make up for the screwups. This was approximately equivalent to maxing out your credit cards by buying lottery tickets.
About the only nice side effect of the tech bust is that a lot of people are willing to look at methods that don't neglect the long term in favor of the short term. These days I only do projects using agile methods like Extreme Programming. People who have been burnt have a much easier time appreciating their focus on sustainability of development.
One of my most clueful clients paid millions for outsourced development during the boom and got a lot of garbage code. There's nothing like writing off a 7-figure investment to make people more careful about their spending.
My advice is this: suck it up, do some hobby programming, build a portfolio of samples (nothing sells a candidate like good sample code), and keep on plugging. You'll have to prove yourself.
That's good advice. I almost never hire people who are fresh out of college. They may know some basics, but after a lot of short solo projects, they rarely know how to work well in a group, how to write code that communicates well to others, or how to write maintainable code.
But I'm not a snob about professional experience; any experience will do, especially when it's backed by code samples. I'm especially impressed by people who put stuff up on the web and develop a user base; there's nothing like dealing with real users to educate programmers about a lot of the things that college can't really teach.
Re:Stop looking for "programming" jobs
on
Exporting Myself?
·
· Score: 2, Interesting
Go for a design job. [...] Who wants to be just a programmer anyway? It's like manual labor for your fingers.
If you think that, you'd be doing the job wrong, whether you're a designer or a programmer. As a consultant who has seen a lot of projects, I find few people more dangerous than self-styled architects who consider themselves too good for coding.
On my last project, it was my great pleasure to recommend that they fire all 20 of their architecture group, and then offer them the opportunity to interview for developer positions on the development teams. In two years of design, they had produced a lot of white papers, a bunch of recommendations, a number of frameworks, and a whole lot of mandates. The code base was hugely overarchitected, tangled, and confused.
Most everything the "architects" produced sounded plausible, and many had good ideas at the core. But almost all of it was useless in practice. Why? Because they never had to deal with the practical consequences of their work. Instead of sitting down with the developers and seeing how their theories worked out, they just stayed in their offices and produced more theories. They were deaf musicians with a captive audience.
I use per-month folders and that's worked pretty well for me; the main benefit being that I actually put the papers in the folders. On the rare occasions I've needed to find some scrap of paper, rummaging through a month or two's worth of stuff isn't too bad.
The only drawback I found was at end of year when I wanted to sort out my business expenses so I could deduct them. My solution: two sets of monthly folders: yellow for business, green for personal.
Also, for those fellow geeks who have trouble keeping organized, I strongly recommend the book Organizing from the Inside Out. It's a smart, practical book that really engages your analytical skills. It's done wonders for me.
That's no fair! It's not my fault that I live in a country where our leaders can't do math and are consequently flushing our currency down the toilet!;-)
the counter for funds received over at Wikimedia is now at $23,382.17
Holy moly! That was fast! Hooray for the Internet!
(For those wanting to check the latest numbers, you can find them on the letter. And there's a nice thank you note that also explains what they'll do with the extra dough.)
So what happens when an expert fixes something, then a non-expert 'unfixes' it to match what they 'know' and what matches popular perceptions
Then the expert realizes that in a document for the whole world, they must respect and engage their audience. They add a comment stating their point of view and engage in dialog to convince the other people maintaining that page.
Eventually a consensus will emerge. The consensus might be view A or view B alone. More likely, it will be some combination, perhaps A+B, or perhaps taking the congruent parts of each and agreeing to disagree for now on the parts that are contentious. This can seem like an insoluble problem too, but it's the same problem science has dealt with pretty well for a few centuries.
In the particular case you describe, where there's a common misperception, a good way to prevent regress is to add a note addressing the misperception. E.g., "Although many feel shaking a Polaroid picture helps it to develop faster, study foo shows it makes no difference. The misperception is thought to arise from cause bar." Including, of course, real links to external studies or other Wiki documents.
The Wiki things are cool in a way, but too filled with unqualified opinion.
You're missing the magic behind Wikis.
Most web pages are static, or if they're dynamic, the reader isn't the one with the power to change things. On a Wiki, anybody can come by and help edit.
That doesn't seem like a big deal, but it's amazingly powerful. When I first used the original Wiki, I notice that one sentence in an otherwise good page was confusingly phrased. And so I fixed it. In a few seconds. Wikis allow you to aggregate small amounts of effort from thousands of people.
If the Wikipedia is currently imperfect, that's ok. As experts come by and look at it, they'll fix things that they notice are wrong. It will never be completely perfect, but that's ok; no document ever is: caveat lector was good advice long before the web.
The advantage of the Wiki is that it's a document with an extremely low cost of change, so that it will be able to stay in sync with current knowledge and viewpoints much better than, say, a paper document like Britannica.
I'm too cheap to donate, and I'm only 16 anyway...
I'm not! I just sent them $100. It's a good resource, and a fascinating experiment in collaborative content generation.
Remember the excitement about the internet circa 1997? Well Amazon turned out to be a big mall, and eBay turned out to be a big flea market. But the Wikipedia is pushing the boundaries of what the web is. Those of you who miss the exitement of the early days should check it out. And send them a check so you can see how it turns out.
As a software designer, I am amazed by Wikis. If somebody asked me to build a system that would allow tens of thousands of people to collaborate on the same big document, I would have come up with something an order of magnitude more complicated than The Wikipedia and two orders of magnitude more complicated than Ward Cunningham's original Wiki. But they work amazingly well. $100 is a small price to pay for what I learned studying and using Wikis.
Fourth, grunt tasks mean increased redundancy in the code base.
You are making assumptions regarding those tasks, and your other comments make assumptions regarding these "grunts." If the tasks are truly "mundane" and "automatable" then why are you assuming they have a high unique-error frequency?
I'm not saying the redundant code will have a higher rate of errors. I'm saying that any redundant code increases development inertia. If you read the literature on copy-and-paste programming, you'll find lots of info on why that's bad. Grunt work, by its nature, means that there will be a lot of redundancy. Otherwise you'd call it creative work.
Or government projects, where a new administration can wipe out a decade of development, taking totally unforeseen directions. Having automated generically for contingencies in such an environment would hvae been a terrible waste.
I'm not advocating premature generalization. That would raise costs uselessly and reduce agility. I'm only advocating increasing the expressiveness of the code base by removing duplication. That by necessity removes opportunities for grunt work. Here's why:
If you can give somebody a little info and have them run off and do a bunch of work, then that work has low information content. A bunch of work with low information content means high redundancy. If you forbid duplication, then that means that your programmers will need more information per unit of time developing. Ergo, it's no longer grunt work; it's teamwork.
I agree that one should carefully examine the cost-benefit tradeoffs surrounding any choice in development. But agile software development, and Extreme Programming in particular, advocate similar things to what I'm saying here. Those projects end up being much more productive than traditional development. I think that's because much thinking in traditional methods about cost/benefit tradeoffs is deeply flawed.
Your efforts to defend your mention of the equivalence of Bush and Saddam in your mind are simply semantics. You said you dislike them equally, so you consider them morally equivalent. Easy enough.
Eh? I dislike Britney Spears more than I dislike Ghengis Khan, but that doesn't mean I consider brutal dictators to be morally superior to pop stars.
You can condemn his father with some justice, but George W Bush is doing the morally correct thing, insofar as I can determine.
Living in San Francisco, I agree that there is certainly a chunk of the American left that is irrationally biased against George Bush to a point that is ridiculous. It seems to be the same thing (if smaller in scale) as the American right's irrational hatred of Clinton. It has a lot to do with their personalities, and very little to do with their capabilities.
However, you shouldn't let that blind you to what are very reasonable critiques of the Bush administration. The war wasn't sold to the public on the basis of removing a bad dictator; it was sold on the basis of Iraq posing an imminent threat. Tony Blair got reamed for this, but the American media gave Bush pretty much a free pass on it, as demonstrated by the large proportion of Americans who still have basic facts wrong (especially, as it shows on page 13, the ones who get most of their news from Fox). Many lefty commentators feel that the Bush administration is directly responsible for this, but at the very least they have done very little to correct these misperceptions.
But even if you believe that military action against Iraq was the only option (and many who oppose Bush felt that way), there are still very reasonable gripes with the way it happened. The unilateralist rush to immediate action spent vast amounts of international political capital. Two years ago the world was saying "We are all New Yorkers now." Recent surveys show a drastic falloff in worldwide approval of America as well as loss of stature for the UN. This isn't just a long-term issue, either; the lack of support from other nations, especially Islamic ones, is making it much harder to turn Iraq into a functioning democracy.
Personally, I favored the use of force to bring Iraq into compliance with UN anti-proliferation treaties. I also favor the use of the US's power to benefit the downtrodden everywhere, especially by furthering the spread of the freedoms that we in western democracies often take for granted. But I think our Iraqi adventure was poorly executed and done for the wrong reasons.
The phrase "Literature Geek" makes me wonder, can you be a "Sports Geek"? Or a "Fashion Geek"?
I know of sports geeks, but I personally know some fashion geeks. It's the same as with a computer geek: they have a deep knowledge of the topic and takes great intellectual delight in studying and practicing it.
One of them I know makes her living from it; she buys up clothing at estate sales and low-end vintages stores and sells it at the high-end ones. The main problem she has is that she wants to keep it all for herself; her house would make a good museum.
And if it makes you feel better, she's also a big science fiction fan. But other fashion geeks I know aren't.
Therefore, there are many cases in which the automation investment is more costly than inexpensive human labour -- humans are still more rapidly and inexpensively trainable than computing devices, for certain tasks.
This is true, but not relevant for progamming teams for at least five reasons.
One, if something is automatable, you should not pay a programmer a lot of money to do it. Hire a drone.
Second, the cost of software development increases exponentially with the size of the project (with some interesting discontinuous leaps). This is because the bottlenecks around software development are related to human communication costs. By relentlessly spending the time (or money) to automate a task, you keep your team size smaller, staying on the low end of the cost curve.
Third, the next most important factor in software project cost is related to managing complexity. Relentless simplification and automation keep that linear (or close to it), not the traditional exponential curve. You know those projects that have to throw away the code base between versions? Those are almost always failures in complexity management, and they are very, very expensive. As Netscape learned.
Fourth, grunt tasks mean increased redundancy in the code base. This means that for equivalent functionality, you'll have much larger code bases unless you invest in automation. This means much higher maintenance costs, and much larger bug counts (because instead of fixing a bug once, you have to find every time a grunt did it). Better to abstract the duplication or automate it away.
Fifth, automation increases agility. Agility allows business to cope with change, and better, allows businesses to create change and force their competitors to cope. If you have to manually test your whole application before each release, you can't release every week, and you'll be lucky to release every few months. This drastically limits the speed and confidence with which you can change core code.
Try using an agile process (like Extreme Programming) for six months, and you'll see what I mean. Automation of routine tasks isn't a luxury; it's the fastest way to high productivity. See for example, this post today on the Extreme Programming mailing list. The last couple paragraphs contain comments from a product manager that says that he's having a hard time coming up with requirements fast enough for the team, and that the team is "exponentially" more productive than teams at competitors.
A lot of people think programming is art or something like it. The question is, should they?
Those interested in the topic should check out Artful Making, a book by a business school professor and a theater prof. The basic argument is that many kinds of knowledge work, including software development, are better suited to methods used in collaborative artistic pursuits (like putting on a play) than they do with industrial activities.
His view is the programming is like plumbing or carpentry.
I think this analogy is correct on the small scale (e.g., a one-person programming job is a lot like a one-person carpentry job). But I think the analogy breaks down when you scale up.
If you are building a 3000-room hotel, it makes sense to use the plan-it-all-out-and-divide-up-the-work methods common in the building industry. But software is fundamentally different than a building, in that buildings tend to be pretty stable, but large software projects are never done. Has anybody here ever heard of a software product that a) has been successfully in use for years, b) is still at v 1.0, and c) isn't trivially small?
I've seen too many colleagues who just wanted programming requests left in their intray and didn't want to work actively with the users. That kind of relationship is easily outsourced, as opposed to the person who understands not just the code but the working process that it supports.
Yes! Absolutely.
Developers should look at agile processes like Extreme Programming. All of them require more collaboration between geeks and non-geeks. XP, for example, requires that a product manager be in the same room as the developers.
Businesses like these processes because they're fast and efficient compared with traditional methods. But the outsourcing trend provides programmers with a great reason to make the effort: it lets them heavily exploit the one advantage over distant programmers: the high-bandwidth, low-latency medium of face-to-face communication.
We offloaded some of our mundane programming tasks to an Indian firm
If you have mundane programming tasks, you're doing it wrong. If something is truly mundane, that means it's automatable. If it's automatable, then since it's already on a computer, you should just go ahead and automate it.
Software projects are tracked and managed. It soon would be apparent that your progress is not aligned with what the initial estimate was, and although you could give some bullshit reasons as to why your progress was not as expected, they would eventually get rid of you for somebody more efficient.
You'd think that, wouldn't you. Obviously, you've never seen one of the large IT consulting firms in action. The way it works for them is that they
make an initial estimate of $5 million for a $500k job;
bring in a bunch of people just out of college, all of whom are good looking but clueless;
appear to work feverishly
write a lot of reports and produce a lot of charts;
get nothing done, and then
charge another $5 million to try again;
and then it's back to step 2.
I'm sure that some people in those companies are neither mostly incompetent or utterly shameless, but I have no personal experience that proves otherwise.
Does anybody make a small box for home router use?
The consumer-grade DSL router appliances all seem lame. I'd like to set up a box running linux with a couple of ethernet ports, so that I can route between DSL, 801.11, and ethernet, with some of my own customizations.
I could get an old PC, but they are large, noisy, power-hungry, and prone to failure. I just want some simple linux appliance with a couple of ethernet ports, 64 MB of RAM, and a couple hundred MHz of processor. It'd be especially nice if it were power-thrifty. E.g., using one of the Transmeta processors.
But I see on any thread involving spammers a sort of reflexive hatred - and assumption of Evil Intent - which would seem more appropriate for a religious war on some Christian board.
Part of this is a hardware issue. Humans (and many other social animals) have built-in hardware for detecting cheaters, and a built-in bias against them. It's understandable that a lot of techies, who understand what's going on and are forced to deal with the problem, get worked up.
Another part of it is that for a lot of us, this was an obvious problem nearly ten years ago. A decade of seeing something important to us rot has made a lot of people frustrated, furious, or bitter. Especially so given that the problem is caused by a relatively small number of greedy idiots, abetted by a lot of people who either lack foresight or think it's somebody else's problem.
Of course, it's not just the spam problem that Americans are having trouble getting their heads around. Anybody who can read the newspapers and do a little math could see that there are huge problems brewing with Social Security, Medicare, health care costs, the federal budget, campaign finance, and congressional districting, just to name a few. It's as if people of all political stripes are putting their hands over their ears and singing show tunes, hoping the problems will just go away.
It'd be nice to blame this on an evil conspiracy making use of S.E.P. Field generators. But I think the actual causes are subtle and poorly understood. I have my own theories, but I'd be interested to hear how Slashdotters think this has come about.
So, I ask, simply - is there any substantial evidence that Spam is truly a threat larger than just being a general annoyance?
Yes. There are reasonable estimates from both US and European organization that the costs of dealing with spam are in the billions of dollars annually. This is the same kind of money that it would take to fund, say, proper anti-AIDS efforts in every poor country in the world.
There are also reliable statistics from a number of organizations that spam is well over 50% of email traffic. There are also stats from multiple sources that spam is still growing exponentially, and at a rate much faster than regular email traffic.
This growth has been going on for years. If you look back at older Slashdot articles about spam, practically every one says, "Gosh, it's not a big problem for me. I just hit delete." For years, I and others here have answered, "Well, if its growth is left unchecked, it will be a problem eventually, even for you." And it will. No matter where you put the threshhold between mere annoyance and real threat, exponential growth will get you there eventually.
what they do is put together the hardware, software, and support in a convenient package for the end-user. As for the morons, well, they pay the bills.
Amen to that. I'm so tired of elitist geeks, especially ones who make unusable garbage and then blame the user for not getting it. Really smart geeks recognize that if they make stuff that only a geek can use, that's about 1% of the population. If they want broad adoption of something, then they have to make it so that people in the 20th percentile can use it, too.
What really gets me is that some of the people who are the worst offenders in this don't seem to realize that they are acting just like the people they love to hate. Those snobby, ignorant jocks and cheerleaders? They're also in the top 1% in a small slice of the world, and they measure everybody else in terms of their particular talents. Annoying and shallow, isn't it?
It would please me greatly if the users-are-all-idiots crowd would use some of their immense brainpower to discover that there are more aspects to people and the broader world than can be measured by IQ tests.
Remember that your online backup provider's backup versions of your files are fair game for subpoenas from the RIAA or whoever else can shell out a few bucks to a lawyer.
Uh, so are your own backup tapes. As Microsoft certainly discovered.
I could imagine that a small online backup company might roll over more quickly than you would. But the opposite case is more plausible to me.
If I were running an online backup company, I'd want people to feel that there data was very safe with me, and part of that would be a deep reluctance to give out data to third parties. Fighting hard against a dubious subpoena would be great advertising.
And if I were suing somebody and wanted their data, I don't see why I'd go after the backup company anyhow. The data, even when the backup company has it, still belongs to the company who had it backed up. It seems to me that I'd have to convince a judge that I had a legitimate beef with the owner of the data; to be fair the judge would have to hear from that owner. Then I'd think the judge would just order the owner to surrender the data, and let them work out where it was kept. I'd only get an order for the backup company if the person I'm suing wasn't coughing up.
So really, the only case where I'd be reluctant to use an on-line backup company was if I were thinking I needed to defy a court order and destroy my backup tapes. But that's pretty dumb, too. Here's my tip to all you criminals out there: if you're doing something bad, don't keep backups of the evidence against you.
Slashdot Mirror
Using IPv6 won't change that. It would technically mean we have an abundance of addresses, but our ISPs would still pull the same BS, expecting us to pay more for the same level of service.
Possibly not.
Back in Ye Olden Days, IP addresses were free and easy to get. But they became a relatively scarce resource, and companies started charging because of that.
When IPv6 takes off, ISPs will be able to give out as many addresses as they like without incurring significant costs. With even a modicum of competition, that will stop them from charging for extra addresses. And it certainly means they won't have to charge you for a single real address. If nothing else, that will result in reduced support calls.
Mind you, that may not apply to fixed addresses. People who need fixed addresses are up to something more than the average consumer. From the business perspective, it makes sense to charge more to them.
No. LEC. Local Exchange Carrier. It's in the FCC tariffs and all.
Sorry, the people I know talk about them collectively as LECs and separately as the ILEC (for incumbent) vs the CLECs. For an example of this usage, see this interesting prediction.
And yes, we pay $1700/month for a local DS3 on top of those fees to carry the circuits to us from the telco.
Wow. That's just robbery. $37.50 per month just to use the pair plus a slice of the DSLAM? Especially given that they also charge a monthly fee for the phone line and would have to maintain the wires anyhow? Ridiculous.
our LEC charges us $37.50/month for line provisioning on each 768/128 circuit
Holy moly! Is that an ILEC or a CLEC? And what do you get for that? Do you guys pay for the lines to the DSLAMs on top of that?
Again, here I am, the sysadmin for SCO. [...] I'm not the devil on Darl's shoulder telling him "Psst, ok, now sue these guys!". [...] would you still pass me by for a less qualified applicant.
Probably. I can teach skills. Fixing your ethical handicaps is beyond me.
SCO is a good example. Spammers are another one. I would expect people who work for them to be ethically challenged. Either that or unaware of what their employers were up to, in which case they'd be too clueless to bother with.
Why again, would hiring a kickass forward thinking ex-SCO sysadmin [...] not be in the "best interests of their firms"
The overly dramatic choice you set up, between the inadequate but nice employee or the skilled former concentration camp guard, isn't the one hiring managers face. The reality is that the manager will end up with a bunch of people who will probably be fine; the challenge is in picking the best one in the long term.
There are a few reasons I'd lean away from somebody with an ethically suspect past.
One would be that ethical problems are hard to detect. If somebody is incompetent, I'll know in pretty short order. If somebody is casing the joint for embezzlement opportunities, I might not know until the bank account is empty. Or, less dramatically, I might not realize that he's really clever about covering up shoddy work.
The second, and probably the biggest for me, would be a concern that the person won't get the big picture. A person who doesn't mind making a living spamming can't have much sympathy for the fucking colossal amount of trouble they are causing for their recipients. Why would that person care any more about the end users of the new company's products?
And the third would be simple CYA. Even if there's a relatively small chance that a person working for a corrupt company is themselves corrupt, what happens if it turns out to be true? Then suddenly not only am I the guy who hired the thief, but I'm the guy who hired the thief who used to the sysadmin for Enron's accounting department. Hindsight will make me look like a chump.
One of the least talked about reasons that everything has fallen apart in the tech sector is the sheer worthlessness of so many of the people in it.
This is very true, and it doesn't apply just to the line programmers.
A gold rush mentality meant many people made a lot of decisions that worked in the short term but screwed them in the long term. The theory at the time was that if things worked you'd get VC money (or go IPO) and have so much dough that you could make up for the screwups. This was approximately equivalent to maxing out your credit cards by buying lottery tickets.
About the only nice side effect of the tech bust is that a lot of people are willing to look at methods that don't neglect the long term in favor of the short term. These days I only do projects using agile methods like Extreme Programming. People who have been burnt have a much easier time appreciating their focus on sustainability of development.
One of my most clueful clients paid millions for outsourced development during the boom and got a lot of garbage code. There's nothing like writing off a 7-figure investment to make people more careful about their spending.
My advice is this: suck it up, do some hobby programming, build a portfolio of samples (nothing sells a candidate like good sample code), and keep on plugging. You'll have to prove yourself.
That's good advice. I almost never hire people who are fresh out of college. They may know some basics, but after a lot of short solo projects, they rarely know how to work well in a group, how to write code that communicates well to others, or how to write maintainable code.
But I'm not a snob about professional experience; any experience will do, especially when it's backed by code samples. I'm especially impressed by people who put stuff up on the web and develop a user base; there's nothing like dealing with real users to educate programmers about a lot of the things that college can't really teach.
Go for a design job. [...] Who wants to be just a programmer anyway? It's like manual labor for your fingers.
If you think that, you'd be doing the job wrong, whether you're a designer or a programmer. As a consultant who has seen a lot of projects, I find few people more dangerous than self-styled architects who consider themselves too good for coding.
On my last project, it was my great pleasure to recommend that they fire all 20 of their architecture group, and then offer them the opportunity to interview for developer positions on the development teams. In two years of design, they had produced a lot of white papers, a bunch of recommendations, a number of frameworks, and a whole lot of mandates. The code base was hugely overarchitected, tangled, and confused.
Most everything the "architects" produced sounded plausible, and many had good ideas at the core. But almost all of it was useless in practice. Why? Because they never had to deal with the practical consequences of their work. Instead of sitting down with the developers and seeing how their theories worked out, they just stayed in their offices and produced more theories. They were deaf musicians with a captive audience.
I use per-month folders and that's worked pretty well for me; the main benefit being that I actually put the papers in the folders. On the rare occasions I've needed to find some scrap of paper, rummaging through a month or two's worth of stuff isn't too bad.
The only drawback I found was at end of year when I wanted to sort out my business expenses so I could deduct them. My solution: two sets of monthly folders: yellow for business, green for personal.
Also, for those fellow geeks who have trouble keeping organized, I strongly recommend the book Organizing from the Inside Out. It's a smart, practical book that really engages your analytical skills. It's done wonders for me.
Nah, nah, got you beat! I sent them EUR150! ;-)
;-)
That's no fair! It's not my fault that I live in a country where our leaders can't do math and are consequently flushing our currency down the toilet!
the counter for funds received over at Wikimedia is now at $23,382.17
Holy moly! That was fast! Hooray for the Internet!
(For those wanting to check the latest numbers, you can find them on the letter. And there's a nice thank you note that also explains what they'll do with the extra dough.)
So what happens when an expert fixes something, then a non-expert 'unfixes' it to match what they 'know' and what matches popular perceptions
Then the expert realizes that in a document for the whole world, they must respect and engage their audience. They add a comment stating their point of view and engage in dialog to convince the other people maintaining that page.
Eventually a consensus will emerge. The consensus might be view A or view B alone. More likely, it will be some combination, perhaps A+B, or perhaps taking the congruent parts of each and agreeing to disagree for now on the parts that are contentious. This can seem like an insoluble problem too, but it's the same problem science has dealt with pretty well for a few centuries.
In the particular case you describe, where there's a common misperception, a good way to prevent regress is to add a note addressing the misperception. E.g., "Although many feel shaking a Polaroid picture helps it to develop faster, study foo shows it makes no difference. The misperception is thought to arise from cause bar." Including, of course, real links to external studies or other Wiki documents.
The Wiki things are cool in a way, but too filled with unqualified opinion.
You're missing the magic behind Wikis.
Most web pages are static, or if they're dynamic, the reader isn't the one with the power to change things. On a Wiki, anybody can come by and help edit.
That doesn't seem like a big deal, but it's amazingly powerful. When I first used the original Wiki, I notice that one sentence in an otherwise good page was confusingly phrased. And so I fixed it. In a few seconds. Wikis allow you to aggregate small amounts of effort from thousands of people.
If the Wikipedia is currently imperfect, that's ok. As experts come by and look at it, they'll fix things that they notice are wrong. It will never be completely perfect, but that's ok; no document ever is: caveat lector was good advice long before the web.
The advantage of the Wiki is that it's a document with an extremely low cost of change, so that it will be able to stay in sync with current knowledge and viewpoints much better than, say, a paper document like Britannica.
I'm too cheap to donate, and I'm only 16 anyway...
I'm not! I just sent them $100. It's a good resource, and a fascinating experiment in collaborative content generation.
Remember the excitement about the internet circa 1997? Well Amazon turned out to be a big mall, and eBay turned out to be a big flea market. But the Wikipedia is pushing the boundaries of what the web is. Those of you who miss the exitement of the early days should check it out. And send them a check so you can see how it turns out.
As a software designer, I am amazed by Wikis. If somebody asked me to build a system that would allow tens of thousands of people to collaborate on the same big document, I would have come up with something an order of magnitude more complicated than The Wikipedia and two orders of magnitude more complicated than Ward Cunningham's original Wiki. But they work amazingly well. $100 is a small price to pay for what I learned studying and using Wikis.
You are making assumptions regarding those tasks, and your other comments make assumptions regarding these "grunts." If the tasks are truly "mundane" and "automatable" then why are you assuming they have a high unique-error frequency?
I'm not saying the redundant code will have a higher rate of errors. I'm saying that any redundant code increases development inertia. If you read the literature on copy-and-paste programming, you'll find lots of info on why that's bad. Grunt work, by its nature, means that there will be a lot of redundancy. Otherwise you'd call it creative work.
Or government projects, where a new administration can wipe out a decade of development, taking totally unforeseen directions. Having automated generically for contingencies in such an environment would hvae been a terrible waste.
I'm not advocating premature generalization. That would raise costs uselessly and reduce agility. I'm only advocating increasing the expressiveness of the code base by removing duplication. That by necessity removes opportunities for grunt work. Here's why:
If you can give somebody a little info and have them run off and do a bunch of work, then that work has low information content. A bunch of work with low information content means high redundancy. If you forbid duplication, then that means that your programmers will need more information per unit of time developing. Ergo, it's no longer grunt work; it's teamwork.
I agree that one should carefully examine the cost-benefit tradeoffs surrounding any choice in development. But agile software development, and Extreme Programming in particular, advocate similar things to what I'm saying here. Those projects end up being much more productive than traditional development. I think that's because much thinking in traditional methods about cost/benefit tradeoffs is deeply flawed.
Your efforts to defend your mention of the equivalence of Bush and Saddam in your mind are simply semantics. You said you dislike them equally, so you consider them morally equivalent. Easy enough.
Eh? I dislike Britney Spears more than I dislike Ghengis Khan, but that doesn't mean I consider brutal dictators to be morally superior to pop stars.
You can condemn his father with some justice, but George W Bush is doing the morally correct thing, insofar as I can determine.
Living in San Francisco, I agree that there is certainly a chunk of the American left that is irrationally biased against George Bush to a point that is ridiculous. It seems to be the same thing (if smaller in scale) as the American right's irrational hatred of Clinton. It has a lot to do with their personalities, and very little to do with their capabilities.
However, you shouldn't let that blind you to what are very reasonable critiques of the Bush administration. The war wasn't sold to the public on the basis of removing a bad dictator; it was sold on the basis of Iraq posing an imminent threat. Tony Blair got reamed for this, but the American media gave Bush pretty much a free pass on it, as demonstrated by the large proportion of Americans who still have basic facts wrong (especially, as it shows on page 13, the ones who get most of their news from Fox). Many lefty commentators feel that the Bush administration is directly responsible for this, but at the very least they have done very little to correct these misperceptions.
But even if you believe that military action against Iraq was the only option (and many who oppose Bush felt that way), there are still very reasonable gripes with the way it happened. The unilateralist rush to immediate action spent vast amounts of international political capital. Two years ago the world was saying "We are all New Yorkers now." Recent surveys show a drastic falloff in worldwide approval of America as well as loss of stature for the UN. This isn't just a long-term issue, either; the lack of support from other nations, especially Islamic ones, is making it much harder to turn Iraq into a functioning democracy.
Personally, I favored the use of force to bring Iraq into compliance with UN anti-proliferation treaties. I also favor the use of the US's power to benefit the downtrodden everywhere, especially by furthering the spread of the freedoms that we in western democracies often take for granted. But I think our Iraqi adventure was poorly executed and done for the wrong reasons.
The phrase "Literature Geek" makes me wonder, can you be a "Sports Geek"? Or a "Fashion Geek"?
I know of sports geeks, but I personally know some fashion geeks. It's the same as with a computer geek: they have a deep knowledge of the topic and takes great intellectual delight in studying and practicing it.
One of them I know makes her living from it; she buys up clothing at estate sales and low-end vintages stores and sells it at the high-end ones. The main problem she has is that she wants to keep it all for herself; her house would make a good museum.
And if it makes you feel better, she's also a big science fiction fan. But other fashion geeks I know aren't.
Therefore, there are many cases in which the automation investment is more costly than inexpensive human labour -- humans are still more rapidly and inexpensively trainable than computing devices, for certain tasks.
This is true, but not relevant for progamming teams for at least five reasons.
One, if something is automatable, you should not pay a programmer a lot of money to do it. Hire a drone.
Second, the cost of software development increases exponentially with the size of the project (with some interesting discontinuous leaps). This is because the bottlenecks around software development are related to human communication costs. By relentlessly spending the time (or money) to automate a task, you keep your team size smaller, staying on the low end of the cost curve.
Third, the next most important factor in software project cost is related to managing complexity. Relentless simplification and automation keep that linear (or close to it), not the traditional exponential curve. You know those projects that have to throw away the code base between versions? Those are almost always failures in complexity management, and they are very, very expensive. As Netscape learned.
Fourth, grunt tasks mean increased redundancy in the code base. This means that for equivalent functionality, you'll have much larger code bases unless you invest in automation. This means much higher maintenance costs, and much larger bug counts (because instead of fixing a bug once, you have to find every time a grunt did it). Better to abstract the duplication or automate it away.
Fifth, automation increases agility. Agility allows business to cope with change, and better, allows businesses to create change and force their competitors to cope. If you have to manually test your whole application before each release, you can't release every week, and you'll be lucky to release every few months. This drastically limits the speed and confidence with which you can change core code.
Try using an agile process (like Extreme Programming) for six months, and you'll see what I mean. Automation of routine tasks isn't a luxury; it's the fastest way to high productivity. See for example, this post today on the Extreme Programming mailing list. The last couple paragraphs contain comments from a product manager that says that he's having a hard time coming up with requirements fast enough for the team, and that the team is "exponentially" more productive than teams at competitors.
A lot of people think programming is art or something like it. The question is, should they?
Those interested in the topic should check out Artful Making, a book by a business school professor and a theater prof. The basic argument is that many kinds of knowledge work, including software development, are better suited to methods used in collaborative artistic pursuits (like putting on a play) than they do with industrial activities.
His view is the programming is like plumbing or carpentry.
I think this analogy is correct on the small scale (e.g., a one-person programming job is a lot like a one-person carpentry job). But I think the analogy breaks down when you scale up.
If you are building a 3000-room hotel, it makes sense to use the plan-it-all-out-and-divide-up-the-work methods common in the building industry. But software is fundamentally different than a building, in that buildings tend to be pretty stable, but large software projects are never done. Has anybody here ever heard of a software product that a) has been successfully in use for years, b) is still at v 1.0, and c) isn't trivially small?
I've seen too many colleagues who just wanted programming requests left in their intray and didn't want to work actively with the users. That kind of relationship is easily outsourced, as opposed to the person who understands not just the code but the working process that it supports.
Yes! Absolutely.
Developers should look at agile processes like Extreme Programming. All of them require more collaboration between geeks and non-geeks. XP, for example, requires that a product manager be in the same room as the developers.
Businesses like these processes because they're fast and efficient compared with traditional methods. But the outsourcing trend provides programmers with a great reason to make the effort: it lets them heavily exploit the one advantage over distant programmers: the high-bandwidth, low-latency medium of face-to-face communication.
an MBA from a decent, fairly reputable Business School WILL take you places, regardless of your skillset
Regardless of skillset? I see you've had to work with a bunch of fresh-out-of-school MBAs, too.
We offloaded some of our mundane programming tasks to an Indian firm
If you have mundane programming tasks, you're doing it wrong. If something is truly mundane, that means it's automatable. If it's automatable, then since it's already on a computer, you should just go ahead and automate it.
You'd think that, wouldn't you. Obviously, you've never seen one of the large IT consulting firms in action. The way it works for them is that they
- make an initial estimate of $5 million for a $500k job;
- bring in a bunch of people just out of college, all of whom are good looking but clueless;
- appear to work feverishly
- write a lot of reports and produce a lot of charts;
- get nothing done, and then
- charge another $5 million to try again;
- and then it's back to step 2.
I'm sure that some people in those companies are neither mostly incompetent or utterly shameless, but I have no personal experience that proves otherwise.Does anybody make a small box for home router use?
The consumer-grade DSL router appliances all seem lame. I'd like to set up a box running linux with a couple of ethernet ports, so that I can route between DSL, 801.11, and ethernet, with some of my own customizations.
I could get an old PC, but they are large, noisy, power-hungry, and prone to failure. I just want some simple linux appliance with a couple of ethernet ports, 64 MB of RAM, and a couple hundred MHz of processor. It'd be especially nice if it were power-thrifty. E.g., using one of the Transmeta processors.
Anybody seen anything like that?
But I see on any thread involving spammers a sort of reflexive hatred - and assumption of Evil Intent - which would seem more appropriate for a religious war on some Christian board.
Part of this is a hardware issue. Humans (and many other social animals) have built-in hardware for detecting cheaters, and a built-in bias against them. It's understandable that a lot of techies, who understand what's going on and are forced to deal with the problem, get worked up.
Another part of it is that for a lot of us, this was an obvious problem nearly ten years ago. A decade of seeing something important to us rot has made a lot of people frustrated, furious, or bitter. Especially so given that the problem is caused by a relatively small number of greedy idiots, abetted by a lot of people who either lack foresight or think it's somebody else's problem.
Of course, it's not just the spam problem that Americans are having trouble getting their heads around. Anybody who can read the newspapers and do a little math could see that there are huge problems brewing with Social Security, Medicare, health care costs, the federal budget, campaign finance, and congressional districting, just to name a few. It's as if people of all political stripes are putting their hands over their ears and singing show tunes, hoping the problems will just go away.
It'd be nice to blame this on an evil conspiracy making use of S.E.P. Field generators. But I think the actual causes are subtle and poorly understood. I have my own theories, but I'd be interested to hear how Slashdotters think this has come about.
So, I ask, simply - is there any substantial evidence that Spam is truly a threat larger than just being a general annoyance?
Yes. There are reasonable estimates from both US and European organization that the costs of dealing with spam are in the billions of dollars annually. This is the same kind of money that it would take to fund, say, proper anti-AIDS efforts in every poor country in the world.
There are also reliable statistics from a number of organizations that spam is well over 50% of email traffic. There are also stats from multiple sources that spam is still growing exponentially, and at a rate much faster than regular email traffic.
This growth has been going on for years. If you look back at older Slashdot articles about spam, practically every one says, "Gosh, it's not a big problem for me. I just hit delete." For years, I and others here have answered, "Well, if its growth is left unchecked, it will be a problem eventually, even for you." And it will. No matter where you put the threshhold between mere annoyance and real threat, exponential growth will get you there eventually.
what they do is put together the hardware, software, and support in a convenient package for the end-user. As for the morons, well, they pay the bills.
Amen to that. I'm so tired of elitist geeks, especially ones who make unusable garbage and then blame the user for not getting it. Really smart geeks recognize that if they make stuff that only a geek can use, that's about 1% of the population. If they want broad adoption of something, then they have to make it so that people in the 20th percentile can use it, too.
What really gets me is that some of the people who are the worst offenders in this don't seem to realize that they are acting just like the people they love to hate. Those snobby, ignorant jocks and cheerleaders? They're also in the top 1% in a small slice of the world, and they measure everybody else in terms of their particular talents. Annoying and shallow, isn't it?
It would please me greatly if the users-are-all-idiots crowd would use some of their immense brainpower to discover that there are more aspects to people and the broader world than can be measured by IQ tests.
Remember that your online backup provider's backup versions of your files are fair game for subpoenas from the RIAA or whoever else can shell out a few bucks to a lawyer.
Uh, so are your own backup tapes. As Microsoft certainly discovered.
I could imagine that a small online backup company might roll over more quickly than you would. But the opposite case is more plausible to me.
If I were running an online backup company, I'd want people to feel that there data was very safe with me, and part of that would be a deep reluctance to give out data to third parties. Fighting hard against a dubious subpoena would be great advertising.
And if I were suing somebody and wanted their data, I don't see why I'd go after the backup company anyhow. The data, even when the backup company has it, still belongs to the company who had it backed up. It seems to me that I'd have to convince a judge that I had a legitimate beef with the owner of the data; to be fair the judge would have to hear from that owner. Then I'd think the judge would just order the owner to surrender the data, and let them work out where it was kept. I'd only get an order for the backup company if the person I'm suing wasn't coughing up.
So really, the only case where I'd be reluctant to use an on-line backup company was if I were thinking I needed to defy a court order and destroy my backup tapes. But that's pretty dumb, too. Here's my tip to all you criminals out there: if you're doing something bad, don't keep backups of the evidence against you.