I found them reasonably helpful when I had problems with the download manager, and they were responsive a couple of years ago when they switched back from requiring the use of their download manager at the request of Linux users like me who complained to them about it. Weirdly, the only way I've ever been able to make the eMusic provided download manager work was through Squid. Without a proxy configured, it always failed to actually download anything. Never figured that one out, and I never saw anyone else with the same problem, so I chalk it up to something weird about my system.
As others have pointed out, there are several other free download managers that work fine on Linux. I used to use the perl one, and it worked very nicely, but now that I've figured out the trick to using the eMusic one, I use it.
I love me some eMusic. It certainly works better with Linux than iTunes or any of the other big ones.
You're just being silly. In my experience yum is easier than apt-get (by a tiny amount), faster than apt-get (often by a large amount, but usually not noticeably so, and there are cases where apt-get is faster), and much smaller than apt-get (I like small software, as it's easier to fix--I once patched an early version of yum to re-add authentication support because I needed it and it took all of two hours to do...I couldn't even begin to grasp the apt-get 200k+ line codebase in two hours...I also suspect there are more bugs in apt-get because there are a lot more lines of codes for bugs to hide in).
To update your system with the latest packages:
yum update
vs.
apt-get update apt-get upgrade
Why the added "apt-get update" step? Because we need to make sure the repository data is up to date. yum checks for us, and downloads new data if the repository has changed. It only downloads new repodata files if the repository has changed, unlike the assertion of another poster that it downloads it all every time you use it. Software ought to do that extra step for me; I'm clearly connected to the network if I'm doing an "upgrade". It doesn't make sense to make the user do an extra step. yum offers the option of listing repos without pulling down repodata, since that can be done without network access. apt-get doesn't offer the choice of automatically getting new repo data. It seems to me that the edge case of listing packages without connecting to the net trumps the common case of updating or installing software in apt-get. That is a small wrongness that bugs me every time I use it.
That said, they both work amazingly well and I love them both. I am perfectly content to use either one on any system I manage. They far surpass yasts package management tools on SUSE, and many systems don't even have anything remotely comparable. I consider a system like yum or apt-get to be a minimal level of package management capability for any server I choose to deploy. Thus I'd never roll out a Mac OS X server, despite the quality of the hardware and shiny-ness of the GUI. Likewise for a Sun machine: until they have a system like yum/apt-get they're not even in the running. Patching on Sun is laughably obtuse, or it was a year or two ago when I last managed a Sun system. At least Windows allows OS updates to be performed easily and with some automation (but not effectively from the command line, and none of the non-MS software can be updated via Windows Update). But I'll happily deploy Fedora or Debian or CentOS or Ubuntu systems for production use. If up2date fully supported yum repositories (including the authentication support I mentioned needing) I would include RHEL in this list. Kickstart also rocks my socks and I hate not having it, but this discussion isn't about automated installations.
Soybeans are quite nutritious and an excellent source of complete proteins and fiber. Like almost any rich and high protein food, they should probably be eaten in moderation because large amounts are potentially harmful to long-term health (reduced mineral absorption seems to be a common theme in available research, but the research is inconclusive), but to imply that any other food produced by American farmers is likely more nutritious than soybeans is simply untrue. Soybeans are among the most nutritious plant foods and the most healthful to eat on a regular basis, reducing the risk of prostate cancer in men and breast cancer in women among many other health benefits.
A quick google turns up a lot more good news about soy as a food source than bad news. The NIH and the AHA both consider soy an extremely beneficial food, and the FDA recommends daily soy consumption. If it is replacing a high fat animal-based protein, the benefits are likely even greater. Just don't eat predominantly soy at every meal, because balance and variety is also necessary for health. Broccoli is also really good for you, but if you're only eating broccoli, after a while things just aren't going to go so well for your health.
a) squandering the funding on absurd levels of growth and infrastructure, b) failing to IPO in time, and (worst of all)
Seems to me that A is pretty much the standard prelude to B in VC-backed explode-o-pop companies. How else did you want them to go about doing an "IPO in time"? Given that it all came down to gambling that you'd get out before the bubble burst back then, I can't help but think that the problem was that the failure was in not building a viable business...I'll certainly concede that it could have been VC bad apples causing the trouble, but I suspect it was their desire to "IPO in time" that led to their bad decisions. In other words, I bet the VC bad apples you mention probably were trying desperately to pull an IPO out of their hats (or wherever was convenient) as fast as possible. VCs come in the door with the words "exit strategy" first, last, and right smack dab in the middle of their minds (and there's nothing wrong with that, as long as you're aware of it). During the boom an IPO was the most viable exit strategy.
Anyway, maybe I'm wrong, but it seems like these A and B statements are at odds with each other, given the time and the economic state of the world in 2001. And C seems to be the most common story of 2001, and it probably comes back to the mindset of getting to IPO in time, rather than building a viable and self-sustaining business (if you're making money faster than you're spending it, the IPO can wait as long as you need it to or not happen at all and you don't have to give up all control of the company to VCs in order to keep running...interesting concept, I know, and one that was foreign to the era). Actually, I probably am wrong. I also had an Open Source business during the 1999-2001 years, and somehow I didn't come out a millionaire...I must have done something wrong. Maybe I didn't IPO fast enough.
So, to sum up, it seems pretty clear to me that Red Hat would still be around and making money even without the successful IPO, while most of the businesses that disappeared, LinuxCare included, would have failed even without the frenzy of the boom/bust to speed things along (or string them out). I wonder if there are any public studies on "making money before IPO or significant VC investment == still in business ten years later"? Seems like there'd be a mighty strong correlation there, and I don't think anything that happened during the boom would throw off that correlation to any significant degree. Sure, sometimes companies limp along for years after a hugely successful IPO (cough!VALinuxcough!) burning through the cash they made by getting to IPO fast enough, but they might be the exception that proves the rule.
Just wanted to chime in on the Hurrah for n-Track thread.
I no longer use n-Track (I don't have a Windows box anymore, even for audio work) but I've never seen anything better. It's got a mostly traditional multi-track feel, but it diverges from that model when it makes sense to do so. I've used a half-dozen far more expensive packages including Pro Tools, and none had the low learning curve of n-Track (effectively nil, if you know how to work in a multi-track studio) or the seeming lack of limitations. Bugs were almost non-existent even in very early versions (I was using it daily at least seven years ago), and the author is astoundingly good at getting fixes in fast. He once rolled a custom version for me with longer track lengths (I needed 90 minutes, as I was mastering a radio show right in the multi-track editor due to budget constraints). Darned good support for something that costs a few bucks, I think it was $32 when I first registered, and it increased over the years by a trifling sum to $49 or $75 depending on version. If I still used Windows I'd still be using n-Track. I've even considered building myself a dedicated audio box with Windows just so I could use n-Track for my current project, but so far I'm sticking with Ardour, Audacity, and a slew of other OSS tools (and enjoying the experimentation, if not the immaturity of the Jack+LADSPA+DSSI+ALSA platform). In short, it's fantastic.
I'll stop rambling now. Just saw mention of my favorite DAW and thought I'd join the chorus.
Real Book is the name of one brand of fake book. They are still in production, and are now "legit" and wholly above-board. In addition to The New Real Book, which is all new selections in new transcriptions, Hal Leonard has recently re-published all of the Real Book series, with minor corrections. Publishers get their cut, the books are cheaper now than back then, and everybody is happy.
"Fake book" is a generic term for any book of simplified musical scores (not necessarily simplified in the sense of being easy to play, as a lot of the songs in jazz fake books are definitely for advanced musicians) that includes the melody, chord symbols, and usually a pretty good representation of the overall structure of the song. The majority of fake book scores fit on one or two facing pages, so that turning pages is minimized. Fake books predate Real Books by many years, but the Real Book folks did a fantastic job of getting good transcriptions onto paper and getting them distributed very widely. The music industry turned a blind eye to this thriving publishing business (mostly) because it was good for their business to have their artists songs being played reasonably accurately all over the world. And, of course, the musicians themselves were using these books on a daily basis and so certainly wouldn't raise a stink about not getting royalties when their own songs appeared in these bibles of Jazz.
The Real Books are still the best of the lot, by far, and I certainly recommend any musician pick up copies of at least the first Real Book. The quality of The New Real Book series is superior, and the supplemental material (like a page of chord spellings and notation conventions) and arrangements are more complete, but the song selection is less exciting to me. The first Real Book has everything you need to be able to fake your way through a whole career as a working musician (which, I guess was the point).
You can, of course, find all of the old ones online in one form or another...they're just too damned useful, and they have such a long history of ignoring copyright law, that copies seem to grow in any cool dark place. I consider $16.50 a small price to pay to get the corrected versions in a nicely bound volume, but there have been times when I didn't have my books and needed a song and I've downloaded and printed it out...and I don't feel even a little guilty about it.
I looked up my office at the corner of 6th and Congress in Austin, Texas and found that the gigantic Frost Bank building (the tallest building in downtown Austin) next door doesn't exist on the map. The Frost building has been there for over a year...So how old is this data?
To quote from the FAQ:
"Satellite images are current, but not real-time."
This seems a bit vague. Does "over a year old" really equal "current"?
I'm not criticizing, just curious. It's wicked cool even if the pictures aren't quite of the present. I just can't check for traffic jams on MoPac yet.
Some (*cough* Allegiance *cough*) are less than truthful and suck big time.
Just so no one assumes this assertion might be dubious based on it coming from an apparently competing CLEC...ALGX really does suck. I had a client here in Austin switch from ALGX to SBC and service got a lot better...and SBC also sucks horribly in their own special way. ALGX was just a joke. Down time could be counted in days per month, and they actually had a cap in their SLA--so if things were really bad and downtime topped something like 72 hours in a month they wouldn't pay any additional refund over the 72 hours. A few times, this actually happened (and why not, it only cost them three days of the 30 we were paying for). Nearly every weekend from Saturday night at midnight until about 6 AM Sunday the network was simply gone, and it wasn't at all odd for the downtime to last until Sunday afternoon.
I simply can't say it emphatically enough: No matter what the price, ALGX is simply not a good deal. They'll promise the moon, and the folks you talk to are all very friendly (including the techs that run the whole hunkering, clanking, ramshackle thing), but when it comes to what you're paying them for, it just isn't there. The Business Class RoadRunner this client upgraded from to an ALGX T1 had been significantly more reliable and faster (and reliability was the reason for the switch off of RoadRunner).
I didn't find it terribly reliable, and more importantly it was mysterious. I hate mysterious software far more than buggy software.
After years of running fine on SCO UNIX, the same configuration and faxmodem was brought over to Linux (I give discounts for converting businesses from SCO to Linux...it's my way of giving something back to the world), and it proved extremely cantankerous. The servers would just stop working on a semi-daily basis. Nothing in the logs. They just stopped. Worse, they refused to start back up until a lock was manually removed (vague recollection here...I had to call tech support to figure out why it wouldn't restart).
After the several of these random stops, several calls to tech support, and several hours of time spent trying to debug an application that didn't provide any information about why it was failing, I replaced it with Hylafax in about half an hour with minimal cursing. I'd never used Hylafax, and though I wouldn't call it easy to use or well documented, I somehow managed to bungle through and get it running without incident. It has been running mostly fine ever since. I have had a couple of hangs where it thinks the fax isn't ready, which required a re-add of the modem--but I got something in the logs and I was able to diagnose the problem in a few minutes by looking at the available information. So, though HylaFax isn't perfect and still manages to fail sometimes, it is less mysterious about it, and so I very strongly prefer it.
There is no feature of VSIfax that we used that HylaFax doesn't have, but there may be some that others find valuable.
The VSIfax people were generally helpful, and I believe you could do far worse in choosing a proprietary software vendor, but if I were to come upon it again, I would plan on replacing it at some point during the job.
It's a definite door-opener, as others have suggested.
I hire development contractors nearly every month, and every single one of them has been chosen because they proved themselves capable by working on interesting, well-designed, projects. It's not the only way to prove it, but any proprietary software on a resume is out of my reach. I can't read the code.
I'm searching now for a part-time jack-of-all-trades to help out in my new office (i.e. telephone support, perl, python, bash, documentation, some basic book-keeping, building/installing servers, etc.) and I'll be much more likely to hire someone with some Open Source projects on their resume. Not only that, but I'll be much more likely to pay them at the higher end of the pay scale.
Open Source developers are great because I can see the code to get a solid grasp on their abilities, see that they truly enjoy the more complex aspects of technology (because why else would they be writing code for no money?), and I can see that they are more likely to want to take an active part in the kind of business I'm building. I'll probably enjoy having them in the office more, too.
That's not to say that all, or even most, companies that are hiring developers, consider all of these things as positive...it's just that I know that Open Source friendly companies are becoming more common, and you're probably more likely to rise to the top of the resume heap at the companies that will be more fun to work for.
Just think about it like this:
Who would you rather work for:
Google or Microsoft? Which one is more likely to hire an Open Source developer? Right!
Extreme examples, I know, but I think the trend holds for the majority of industries that develop software. I could be wrong, of course, but every time I try to think of a comparison of predominantly proprietary vs. predomintantly OSS, I come up thinking the OSS company would be more fun to work for. Zope vs. Vignette? Red Hat vs. SCO (pre-lawsuits, even)?
Put your projects on the resume. Be proud of your OSS work. Even if the company you're hoping to sign on with doesn't do Open Source, taking pride in your work always looks good to potential employers.
Too late to short SCOX, at least at my broker (Ameritrade), as it has to be trading above five bucks. That's been a long time gone. I'm short in a very small position, but when I added money to my account a few months ago I was unable to sell anymore as it was just under $5. I kept hoping for an uptick so I could short some more, but that just isn't going to happen. Oh, well.
I'll just chime in with a "Me too!" on the "NetGear VPN sucks" issue.
I've always liked NetGear switches and wireless routers. They tend to work well, and I've never had one die out of dozens in the field (that's not to say they're necessarily more reliable than other decent low-end brands, but I've a lot of others prove historically less reliable). But their VPN Routers are atrocious. Simply terrible. They offer a "ProSafe" IPSec VPN client package for Windows that is terribly buggy, confusing to use, and not worth the CD it's distributed on. The router itself is also clumsy to configure and doesn't support road warrior configurations in any sane manner.
We eventually got it working, but it was so clumsy that we switched to a PPTP VPN instead (the server being Linux or Windows 2k3, as available). It may not be as secure as IPSec, but it works for all of the clients easily and I don't have to spend hours on the phone with every person that needs access.
Either way, these machines aren't targetted at you, but at people who's only route to buying a PC is this one.
It doesn't matter who's gaining from what, if someone buys one of these PCs and takes it home as a toy, they've deprived that PC from going to someone who actually NEEDS it, not WANTS it.
Just as someone bought up all of the brains capable of comprehending basic economics before you came along? Google "economies of scale", and then come on back to Slashdot after you have something sane to say.
To put it more simply: You're assertion is idiotic. Actually, now that I think about it, I suppose I've been trolled since no one on slashdot is really that stupid...are they?
It's free (as in beer and speech), integrates cleanly with Webmin and Usermin for system administration and user level access to email, spamassassin configuration, etc.
It doesn't do everything that all of the control panels do, but none of them can do everything that Webmin/Usermin/Virtualmin can do either, so it all comes out in the wash. It has the benefit of being easily modified without violating any licenses and is very polite to the underlying OS and services. Unlike pretty much all of the other products in this space (except for the Positive Software CP+ product that is based off of Virtualmin and Webmin) it does not screw up the configuration files, or require you to do all of the configuration of the services from within the GUI. You can switch between vi or emacs and Virtualmin without fear of losing comments or configuration file order.
It has a couple of long-term commercial sponsors, including my company (Swell Technology) and the previously mentioned Positive Software, so it gets consistent developer attention and fast bug fixes. And like Webmin on top of which it runs, it is easily themeable, provides online help features, and is internationalized with many translations.
It may not be perfect for every purpose today, but it gets better all the time and is already in use at a lot of ISPs and hosting providers, as well as ASPs. Worth a look, at least. You may find that with some customizations (which you can do, or you can hire Jamie or someone else to do) Virtualmin can provide a better fit for less money over the long term than any of the proprietary options.
And since it is free, it will cost you nothing but a few minutes or hours to give it a try.
Re:Except Animals are more likely to be right.
on
Good Bad Attitude
·
· Score: 4, Funny
I'm a hacker...
...Just like many people didn't when the whole Nazi thing was going on in the beginning.
The Three Day Novel Writing Contest has a $50 entrance fee, and old-fashioned submission requirements (paper!). Somehow it seems less enthusiasm, and more clever marketing. And, as the NaNoWriMo FAQ points out:
Did you know there is a group in Vancouver that writes novels in a weekend? Yes, and they are fools. Everyone knows that any deep and lasting work of art takes an entire month to make.
And, I think, the big difference is the social aspect of the NaNoWriMo...I think some of us nerds could do with a little more socializing (I could, anyway). So, while the 3 Day folks did it first (and good on 'em for doing it), the NaNoWriMo has a much stronger pull for me. And I laughed a lot more when reading the website than when I browsed the 3Day site. Funny goes well with foolish confidence.
Someone else mentioned it, but I thought it worth a bit more verbiage.
SQL-Ledger is a full accounting system, and includes purchase orders, quotations, RFQs, sales orders, etc. It also handles general double-entry bookkeeping and most everything else you need to keep up with your money, your suppliers, and your customers. It has nice LaTeX templates for checks, POs, quotes, RFQs, invoices, etc. which are easy to modify if you know LaTeX and not horribly difficult if you don't, since there are instructions for editing with LyX.
I first tried it because of two years worth of failed attempts to make GnuCash work reliably (I've been told GnuCash is more stable today than two years ago when I last tried it), and I really needed to get out of the habit of doing no book-keeping until tax-time or until I realized a customer was late with payment. SQL-Ledger was a last resort because I thought I wanted a local GUI app, so I bounced around trying all kinds of things...and finally settled on SQL-Ledger because it didn't crash, lose data, or lack specific features I needed.
It's written in Perl, reasonably easy to modify, and support from the author is cheap. So, while it's not perfect, I don't hesitate to recommend it to folks who want something a little more flexible than QuickBooks (and the fact that it is web-based has been such a huge win for me that I'll never go back to local GUI apps for this purpose).
Anyway, it's not specific to Purchase Orders, but it handles POs very nicely, and I find I use the bits I understand and leave the rest for later (I'm not much of an accountant, but I learn more every day), and it deals well with working in that way.
OS's make their real money (except for MS's) after it is put into maintence mode.
For me to buy this, I'm going to have to see some concrete examples. Got any links?
Seems like Solaris has made money for Sun without being in maintenance mode. Same for MS. Same for Red Hat, Wind River, QNX, Palm, IBM (who have made more money and lost more money on operating systems than just about anybody), etc...
I have nothing to back up my statements other than vague assertions but then you haven't presented anything other than vague assertions either, so I reckon that's fair.
We use roundup, with some minor customizations. It works very nicely, is somewhat easy to customize, integrates nicely with email, and it is free.
http://roundup.sf.net
My next customization will probably be some mechanism for techs to check "resolve and add to knowledge base" and have it dump out the discussion to a Blosxom format blog entry with a selected category. Then it becomes searchable using either one of Blosxom's search plugins or our current htdig search function (roundup supports search, but our customers don't go there unless they plan to report a problem--and resolved issues don't show up in a search by default...the idea behind roundup is to fix problems and then have them go away, not just talk about them). Blosxom with some minor mods will make a great knowledge base, IMHO, and since the format is dead simple to generate from just about anything it would be pretty easy to make it all work together.
All of which requires, hiring a lawyer. High-sounding principles get you mod points, but they don't pay your legal bills.
I'd like to agree and disagree with this sentiment.
I agree because you should never claim "you'll be hearing from my lawyer". You should/start/ with them hearing from your lawyer, because otherwise it is an empty threat (which they'll guess) and you're very likely to say something that can seriously hurt your case and reduce your chance of winning and even if you do win, your resultant damages.
However, a letter from a lawyer is not horribly expensive. If you draft a very precise letter in advance, and ask them to look over it at their normal rate, and correct any issues, you can get in and out in 1-2 hours. It would be likewise a couple of hours or less if you go to a lawyer that deals in trademark law regularly--he'll have a bunch of pre-fab letters and his secretary will pick the right one for your circumstances. Yes, that's $100-$400 (lawyer rates vary, and how much time they'll want to bill will vary--call around), but you're very likely to nip the problem in the bud and never hear from them again.
If they choose to continue to harass you, your lawyer (yes, now you really have a lawyer and when you wave around "you'll hear from my lawyer" they'll know you mean it), will likely take the case at reasonable rates or on contingency. If you have proof of using the mark for any period of time before they used it or claimed it, or if you are clearly in an unrelated field (or both), then the case is open and shut and few lawyers would turn it down.
The good news, however, is that no lawyer would come after you under such circumstances because the case is unwinnable, if they know you already have a lawyer and are not afraid to use it.
Responding to them personally is the worst thing you can do. Ignore them with prejudice, or have a lawyer send them a letter. Do not respond yourself unless you have training in trademark law, and even then you probably shouldn't respond yourself.
Stupid is so cool that we (meaning you all who voted for him) elected an intelligence underachiever (Bush) rather than a brainiac (Gore).
I agree with your theory, but your example is not a very good one. Gore is also an idiot, though distinctly less of an idiot than our current idiot-in-chief. I can't think of the last intellectually impressive president or presidential candidate we've had. Kennedy, certainly. Nixon, possibly (though paranoia has a negative impact on the clarity of ones thoughts), just to cover both sides of the right-left political spectrum. And though they both were seemingly quite smart, they still managed to do the wrong thing on so many counts it is embarassing, with Vietnam producing many of their biggest mistakes.
I believe America is so far down the path of deifying stupidity that we don't even get viable candidates that have half a brain, much less elect them. I'd like to think that there are lots of really smart people behind the curtains actually making the big decisions (like Greenspan in the Federal Reserve), but the behavior of our government pokes holes right through that theory.
It's disappointing, but par for the course, I guess. Large groups of people acting together are capable of stupidity on a massive scale. It shouldn't be surprising that they would elect a mindless jackass to lead the charge. It is sad to live in a time when the stupidest man to ever hold the office of President might just get reelected. At least the first time through, folks might have been able to claim ignorance...but this time around, everyone knows he's dumber than a box of rocks, and he's still polling in the lead.
And I kinda thought the internet and the dot-com boom would give nerds an edge in the political process...but I suppose we still actually have to get involved in the political process, which isn't nearly as much fun as hacking machines.
Mr. Shyamalan has a real gift for creating moods. He should consider moving to France and revitalizing the French film industry. No seriously. They can't afford the huge Hollywood budgets and have never recovered from the death of Francois Truffaut. He could become a great director like Erich Rohmer there instead of being forced to make stupid films due to Hollywood mass-appeal requirements.
No, seriously. Shyamalan is a jackass who makes really horrible movies (Sixth Sense being a surprising exception--I have to wonder if he had help from an uncredited smarter and more interesting friend).
French film, on the other hand, is actually quite healthy these days. Jean-Pierre Jeunet (City of Lost Children, Amelie), Krzysztof Kieslowski (Trois couleurs, though he died in '96 and didn't work exclusively in France, he's worth watching), François Ozon (Swimming Pool), and the list goes on. A lot of really great films are coming out of France, and overall I've found I enjoy a higher percentage of the French films I see than American films. Perhaps it is merely a filtering effect at work, and only the really exceptional French films see US releases.
It would be nice if Shyamalan would get filtered somewhere on the path to making his next film, so I don't even have to see the embarrassing previews.
Re:Not especially funny, but might be useful
on
Fun With Passwords?
·
· Score: 4, Insightful
I once read a tip about website passwords where you shouldn't have the same password for all sites that need a logic. One of the best suggestions I read was to have a password of say 4 characters, and intersperse the website name into it.
Ah, I get it. So that when the slashdot guys read your password 1s2l3a4s, they'll be able to guess that your yahoo mail password is 1y2a3h4o. Perfect! It's a system whereby you use a different password on every site, but still manage to make it available to anyone who has one of your passwords. In-genius. Really. I mean it.
Slashdot Mirror
I found them reasonably helpful when I had problems with the download manager, and they were responsive a couple of years ago when they switched back from requiring the use of their download manager at the request of Linux users like me who complained to them about it. Weirdly, the only way I've ever been able to make the eMusic provided download manager work was through Squid. Without a proxy configured, it always failed to actually download anything. Never figured that one out, and I never saw anyone else with the same problem, so I chalk it up to something weird about my system.
As others have pointed out, there are several other free download managers that work fine on Linux. I used to use the perl one, and it worked very nicely, but now that I've figured out the trick to using the eMusic one, I use it.
I love me some eMusic. It certainly works better with Linux than iTunes or any of the other big ones.
You're just being silly. In my experience yum is easier than apt-get (by a tiny amount), faster than apt-get (often by a large amount, but usually not noticeably so, and there are cases where apt-get is faster), and much smaller than apt-get (I like small software, as it's easier to fix--I once patched an early version of yum to re-add authentication support because I needed it and it took all of two hours to do...I couldn't even begin to grasp the apt-get 200k+ line codebase in two hours...I also suspect there are more bugs in apt-get because there are a lot more lines of codes for bugs to hide in).
To update your system with the latest packages:
yum update
vs.
apt-get update
apt-get upgrade
Why the added "apt-get update" step? Because we need to make sure the repository data is up to date. yum checks for us, and downloads new data if the repository has changed. It only downloads new repodata files if the repository has changed, unlike the assertion of another poster that it downloads it all every time you use it. Software ought to do that extra step for me; I'm clearly connected to the network if I'm doing an "upgrade". It doesn't make sense to make the user do an extra step. yum offers the option of listing repos without pulling down repodata, since that can be done without network access. apt-get doesn't offer the choice of automatically getting new repo data. It seems to me that the edge case of listing packages without connecting to the net trumps the common case of updating or installing software in apt-get. That is a small wrongness that bugs me every time I use it.
That said, they both work amazingly well and I love them both. I am perfectly content to use either one on any system I manage. They far surpass yasts package management tools on SUSE, and many systems don't even have anything remotely comparable. I consider a system like yum or apt-get to be a minimal level of package management capability for any server I choose to deploy. Thus I'd never roll out a Mac OS X server, despite the quality of the hardware and shiny-ness of the GUI. Likewise for a Sun machine: until they have a system like yum/apt-get they're not even in the running. Patching on Sun is laughably obtuse, or it was a year or two ago when I last managed a Sun system. At least Windows allows OS updates to be performed easily and with some automation (but not effectively from the command line, and none of the non-MS software can be updated via Windows Update). But I'll happily deploy Fedora or Debian or CentOS or Ubuntu systems for production use. If up2date fully supported yum repositories (including the authentication support I mentioned needing) I would include RHEL in this list. Kickstart also rocks my socks and I hate not having it, but this discussion isn't about automated installations.
Soybeans are quite nutritious and an excellent source of complete proteins and fiber. Like almost any rich and high protein food, they should probably be eaten in moderation because large amounts are potentially harmful to long-term health (reduced mineral absorption seems to be a common theme in available research, but the research is inconclusive), but to imply that any other food produced by American farmers is likely more nutritious than soybeans is simply untrue. Soybeans are among the most nutritious plant foods and the most healthful to eat on a regular basis, reducing the risk of prostate cancer in men and breast cancer in women among many other health benefits.
A quick google turns up a lot more good news about soy as a food source than bad news. The NIH and the AHA both consider soy an extremely beneficial food, and the FDA recommends daily soy consumption. If it is replacing a high fat animal-based protein, the benefits are likely even greater. Just don't eat predominantly soy at every meal, because balance and variety is also necessary for health. Broccoli is also really good for you, but if you're only eating broccoli, after a while things just aren't going to go so well for your health.
a) squandering the funding on absurd levels of growth and infrastructure, b) failing to IPO in time, and (worst of all)
Seems to me that A is pretty much the standard prelude to B in VC-backed explode-o-pop companies. How else did you want them to go about doing an "IPO in time"? Given that it all came down to gambling that you'd get out before the bubble burst back then, I can't help but think that the problem was that the failure was in not building a viable business...I'll certainly concede that it could have been VC bad apples causing the trouble, but I suspect it was their desire to "IPO in time" that led to their bad decisions. In other words, I bet the VC bad apples you mention probably were trying desperately to pull an IPO out of their hats (or wherever was convenient) as fast as possible. VCs come in the door with the words "exit strategy" first, last, and right smack dab in the middle of their minds (and there's nothing wrong with that, as long as you're aware of it). During the boom an IPO was the most viable exit strategy.
Anyway, maybe I'm wrong, but it seems like these A and B statements are at odds with each other, given the time and the economic state of the world in 2001. And C seems to be the most common story of 2001, and it probably comes back to the mindset of getting to IPO in time, rather than building a viable and self-sustaining business (if you're making money faster than you're spending it, the IPO can wait as long as you need it to or not happen at all and you don't have to give up all control of the company to VCs in order to keep running...interesting concept, I know, and one that was foreign to the era). Actually, I probably am wrong. I also had an Open Source business during the 1999-2001 years, and somehow I didn't come out a millionaire...I must have done something wrong. Maybe I didn't IPO fast enough.
So, to sum up, it seems pretty clear to me that Red Hat would still be around and making money even without the successful IPO, while most of the businesses that disappeared, LinuxCare included, would have failed even without the frenzy of the boom/bust to speed things along (or string them out). I wonder if there are any public studies on "making money before IPO or significant VC investment == still in business ten years later"? Seems like there'd be a mighty strong correlation there, and I don't think anything that happened during the boom would throw off that correlation to any significant degree. Sure, sometimes companies limp along for years after a hugely successful IPO (cough!VALinuxcough!) burning through the cash they made by getting to IPO fast enough, but they might be the exception that proves the rule.
Just wanted to chime in on the Hurrah for n-Track thread.
I no longer use n-Track (I don't have a Windows box anymore, even for audio work) but I've never seen anything better. It's got a mostly traditional multi-track feel, but it diverges from that model when it makes sense to do so. I've used a half-dozen far more expensive packages including Pro Tools, and none had the low learning curve of n-Track (effectively nil, if you know how to work in a multi-track studio) or the seeming lack of limitations. Bugs were almost non-existent even in very early versions (I was using it daily at least seven years ago), and the author is astoundingly good at getting fixes in fast. He once rolled a custom version for me with longer track lengths (I needed 90 minutes, as I was mastering a radio show right in the multi-track editor due to budget constraints). Darned good support for something that costs a few bucks, I think it was $32 when I first registered, and it increased over the years by a trifling sum to $49 or $75 depending on version. If I still used Windows I'd still be using n-Track. I've even considered building myself a dedicated audio box with Windows just so I could use n-Track for my current project, but so far I'm sticking with Ardour, Audacity, and a slew of other OSS tools (and enjoying the experimentation, if not the immaturity of the Jack+LADSPA+DSSI+ALSA platform). In short, it's fantastic.
I'll stop rambling now. Just saw mention of my favorite DAW and thought I'd join the chorus.
Real Book is the name of one brand of fake book. They are still in production, and are now "legit" and wholly above-board. In addition to The New Real Book, which is all new selections in new transcriptions, Hal Leonard has recently re-published all of the Real Book series, with minor corrections. Publishers get their cut, the books are cheaper now than back then, and everybody is happy.
"Fake book" is a generic term for any book of simplified musical scores (not necessarily simplified in the sense of being easy to play, as a lot of the songs in jazz fake books are definitely for advanced musicians) that includes the melody, chord symbols, and usually a pretty good representation of the overall structure of the song. The majority of fake book scores fit on one or two facing pages, so that turning pages is minimized. Fake books predate Real Books by many years, but the Real Book folks did a fantastic job of getting good transcriptions onto paper and getting them distributed very widely. The music industry turned a blind eye to this thriving publishing business (mostly) because it was good for their business to have their artists songs being played reasonably accurately all over the world. And, of course, the musicians themselves were using these books on a daily basis and so certainly wouldn't raise a stink about not getting royalties when their own songs appeared in these bibles of Jazz.
The Real Books are still the best of the lot, by far, and I certainly recommend any musician pick up copies of at least the first Real Book. The quality of The New Real Book series is superior, and the supplemental material (like a page of chord spellings and notation conventions) and arrangements are more complete, but the song selection is less exciting to me. The first Real Book has everything you need to be able to fake your way through a whole career as a working musician (which, I guess was the point).
Here's the link:
The Real Book at Amazon
You can, of course, find all of the old ones online in one form or another...they're just too damned useful, and they have such a long history of ignoring copyright law, that copies seem to grow in any cool dark place. I consider $16.50 a small price to pay to get the corrected versions in a nicely bound volume, but there have been times when I didn't have my books and needed a song and I've downloaded and printed it out...and I don't feel even a little guilty about it.
I looked up my office at the corner of 6th and Congress in Austin, Texas and found that the gigantic Frost Bank building (the tallest building in downtown Austin) next door doesn't exist on the map. The Frost building has been there for over a year...So how old is this data?
To quote from the FAQ:
"Satellite images are current, but not real-time."
This seems a bit vague. Does "over a year old" really equal "current"?
I'm not criticizing, just curious. It's wicked cool even if the pictures aren't quite of the present. I just can't check for traffic jams on MoPac yet.
Some (*cough* Allegiance *cough*) are less than truthful and suck big time.
Just so no one assumes this assertion might be dubious based on it coming from an apparently competing CLEC...ALGX really does suck. I had a client here in Austin switch from ALGX to SBC and service got a lot better...and SBC also sucks horribly in their own special way. ALGX was just a joke. Down time could be counted in days per month, and they actually had a cap in their SLA--so if things were really bad and downtime topped something like 72 hours in a month they wouldn't pay any additional refund over the 72 hours. A few times, this actually happened (and why not, it only cost them three days of the 30 we were paying for). Nearly every weekend from Saturday night at midnight until about 6 AM Sunday the network was simply gone, and it wasn't at all odd for the downtime to last until Sunday afternoon.
I simply can't say it emphatically enough: No matter what the price, ALGX is simply not a good deal. They'll promise the moon, and the folks you talk to are all very friendly (including the techs that run the whole hunkering, clanking, ramshackle thing), but when it comes to what you're paying them for, it just isn't there. The Business Class RoadRunner this client upgraded from to an ALGX T1 had been significantly more reliable and faster (and reliability was the reason for the switch off of RoadRunner).
I didn't find it terribly reliable, and more importantly it was mysterious. I hate mysterious software far more than buggy software.
After years of running fine on SCO UNIX, the same configuration and faxmodem was brought over to Linux (I give discounts for converting businesses from SCO to Linux...it's my way of giving something back to the world), and it proved extremely cantankerous. The servers would just stop working on a semi-daily basis. Nothing in the logs. They just stopped. Worse, they refused to start back up until a lock was manually removed (vague recollection here...I had to call tech support to figure out why it wouldn't restart).
After the several of these random stops, several calls to tech support, and several hours of time spent trying to debug an application that didn't provide any information about why it was failing, I replaced it with Hylafax in about half an hour with minimal cursing. I'd never used Hylafax, and though I wouldn't call it easy to use or well documented, I somehow managed to bungle through and get it running without incident. It has been running mostly fine ever since. I have had a couple of hangs where it thinks the fax isn't ready, which required a re-add of the modem--but I got something in the logs and I was able to diagnose the problem in a few minutes by looking at the available information. So, though HylaFax isn't perfect and still manages to fail sometimes, it is less mysterious about it, and so I very strongly prefer it.
There is no feature of VSIfax that we used that HylaFax doesn't have, but there may be some that others find valuable.
The VSIfax people were generally helpful, and I believe you could do far worse in choosing a proprietary software vendor, but if I were to come upon it again, I would plan on replacing it at some point during the job.
It's a definite door-opener, as others have suggested.
I hire development contractors nearly every month, and every single one of them has been chosen because they proved themselves capable by working on interesting, well-designed, projects. It's not the only way to prove it, but any proprietary software on a resume is out of my reach. I can't read the code.
I'm searching now for a part-time jack-of-all-trades to help out in my new office (i.e. telephone support, perl, python, bash, documentation, some basic book-keeping, building/installing servers, etc.) and I'll be much more likely to hire someone with some Open Source projects on their resume. Not only that, but I'll be much more likely to pay them at the higher end of the pay scale.
Open Source developers are great because I can see the code to get a solid grasp on their abilities, see that they truly enjoy the more complex aspects of technology (because why else would they be writing code for no money?), and I can see that they are more likely to want to take an active part in the kind of business I'm building. I'll probably enjoy having them in the office more, too.
That's not to say that all, or even most, companies that are hiring developers, consider all of these things as positive...it's just that I know that Open Source friendly companies are becoming more common, and you're probably more likely to rise to the top of the resume heap at the companies that will be more fun to work for.
Just think about it like this:
Who would you rather work for:
Google or Microsoft? Which one is more likely to hire an Open Source developer? Right!
Extreme examples, I know, but I think the trend holds for the majority of industries that develop software. I could be wrong, of course, but every time I try to think of a comparison of predominantly proprietary vs. predomintantly OSS, I come up thinking the OSS company would be more fun to work for. Zope vs. Vignette? Red Hat vs. SCO (pre-lawsuits, even)?
Put your projects on the resume. Be proud of your OSS work. Even if the company you're hoping to sign on with doesn't do Open Source, taking pride in your work always looks good to potential employers.
Too late to short SCOX, at least at my broker (Ameritrade), as it has to be trading above five bucks. That's been a long time gone. I'm short in a very small position, but when I added money to my account a few months ago I was unable to sell anymore as it was just under $5. I kept hoping for an uptick so I could short some more, but that just isn't going to happen. Oh, well.
I'll just chime in with a "Me too!" on the "NetGear VPN sucks" issue.
I've always liked NetGear switches and wireless routers. They tend to work well, and I've never had one die out of dozens in the field (that's not to say they're necessarily more reliable than other decent low-end brands, but I've a lot of others prove historically less reliable). But their VPN Routers are atrocious. Simply terrible. They offer a "ProSafe" IPSec VPN client package for Windows that is terribly buggy, confusing to use, and not worth the CD it's distributed on. The router itself is also clumsy to configure and doesn't support road warrior configurations in any sane manner.
We eventually got it working, but it was so clumsy that we switched to a PPTP VPN instead (the server being Linux or Windows 2k3, as available). It may not be as secure as IPSec, but it works for all of the clients easily and I don't have to spend hours on the phone with every person that needs access.
Either way, these machines aren't targetted at you, but at people who's only route to buying a PC is this one.
It doesn't matter who's gaining from what, if someone buys one of these PCs and takes it home as a toy, they've deprived that PC from going to someone who actually NEEDS it, not WANTS it.
Just as someone bought up all of the brains capable of comprehending basic economics before you came along? Google "economies of scale", and then come on back to Slashdot after you have something sane to say.
To put it more simply: You're assertion is idiotic. Actually, now that I think about it, I suppose I've been trolled since no one on slashdot is really that stupid...are they?
Slack is easy to customize. You edit one line in one file.
/etc directory. Shouldn't be all that hard to pick the right line out of the right file.
Yep. And there's only 62930 lines in 248 files in my
Someone else mentioned it in passing, but since I had a hand in making it happen and I think it's very nice, I'll mention Virtualmin.
Virtualmin.com
It's free (as in beer and speech), integrates cleanly with Webmin and Usermin for system administration and user level access to email, spamassassin configuration, etc.
It doesn't do everything that all of the control panels do, but none of them can do everything that Webmin/Usermin/Virtualmin can do either, so it all comes out in the wash. It has the benefit of being easily modified without violating any licenses and is very polite to the underlying OS and services. Unlike pretty much all of the other products in this space (except for the Positive Software CP+ product that is based off of Virtualmin and Webmin) it does not screw up the configuration files, or require you to do all of the configuration of the services from within the GUI. You can switch between vi or emacs and Virtualmin without fear of losing comments or configuration file order.
It has a couple of long-term commercial sponsors, including my company (Swell Technology) and the previously mentioned Positive Software, so it gets consistent developer attention and fast bug fixes. And like Webmin on top of which it runs, it is easily themeable, provides online help features, and is internationalized with many translations.
It may not be perfect for every purpose today, but it gets better all the time and is already in use at a lot of ISPs and hosting providers, as well as ASPs. Worth a look, at least. You may find that with some customizations (which you can do, or you can hire Jamie or someone else to do) Virtualmin can provide a better fit for less money over the long term than any of the proprietary options.
And since it is free, it will cost you nothing but a few minutes or hours to give it a try.
Hacker's know Godwin's Law.
The Three Day Novel Writing Contest has a $50 entrance fee, and old-fashioned submission requirements (paper!). Somehow it seems less enthusiasm, and more clever marketing. And, as the NaNoWriMo FAQ points out:
Did you know there is a group in Vancouver that writes novels in a weekend?
Yes, and they are fools. Everyone knows that any deep and lasting work of art takes an entire month to make.
And, I think, the big difference is the social aspect of the NaNoWriMo...I think some of us nerds could do with a little more socializing (I could, anyway). So, while the 3 Day folks did it first (and good on 'em for doing it), the NaNoWriMo has a much stronger pull for me. And I laughed a lot more when reading the website than when I browsed the 3Day site. Funny goes well with foolish confidence.
Someone else mentioned it, but I thought it worth a bit more verbiage.
SQL-Ledger is a full accounting system, and includes purchase orders, quotations, RFQs, sales orders, etc. It also handles general double-entry bookkeeping and most everything else you need to keep up with your money, your suppliers, and your customers. It has nice LaTeX templates for checks, POs, quotes, RFQs, invoices, etc. which are easy to modify if you know LaTeX and not horribly difficult if you don't, since there are instructions for editing with LyX.
I first tried it because of two years worth of failed attempts to make GnuCash work reliably (I've been told GnuCash is more stable today than two years ago when I last tried it), and I really needed to get out of the habit of doing no book-keeping until tax-time or until I realized a customer was late with payment. SQL-Ledger was a last resort because I thought I wanted a local GUI app, so I bounced around trying all kinds of things...and finally settled on SQL-Ledger because it didn't crash, lose data, or lack specific features I needed.
It's written in Perl, reasonably easy to modify, and support from the author is cheap. So, while it's not perfect, I don't hesitate to recommend it to folks who want something a little more flexible than QuickBooks (and the fact that it is web-based has been such a huge win for me that I'll never go back to local GUI apps for this purpose).
Anyway, it's not specific to Purchase Orders, but it handles POs very nicely, and I find I use the bits I understand and leave the rest for later (I'm not much of an accountant, but I learn more every day), and it deals well with working in that way.
OS's make their real money (except for MS's) after it is put into maintence mode.
For me to buy this, I'm going to have to see some concrete examples. Got any links?
Seems like Solaris has made money for Sun without being in maintenance mode. Same for MS. Same for Red Hat, Wind River, QNX, Palm, IBM (who have made more money and lost more money on operating systems than just about anybody), etc...
I have nothing to back up my statements other than vague assertions but then you haven't presented anything other than vague assertions either, so I reckon that's fair.
We use roundup, with some minor customizations. It works very nicely, is somewhat easy to customize, integrates nicely with email, and it is free.
http://roundup.sf.net
My next customization will probably be some mechanism for techs to check "resolve and add to knowledge base" and have it dump out the discussion to a Blosxom format blog entry with a selected category. Then it becomes searchable using either one of Blosxom's search plugins or our current htdig search function (roundup supports search, but our customers don't go there unless they plan to report a problem--and resolved issues don't show up in a search by default...the idea behind roundup is to fix problems and then have them go away, not just talk about them). Blosxom with some minor mods will make a great knowledge base, IMHO, and since the format is dead simple to generate from just about anything it would be pretty easy to make it all work together.
All of which requires, hiring a lawyer. High-sounding principles get you mod points, but they don't pay your legal bills.
/start/ with them hearing from your lawyer, because otherwise it is an empty threat (which they'll guess) and you're very likely to say something that can seriously hurt your case and reduce your chance of winning and even if you do win, your resultant damages.
I'd like to agree and disagree with this sentiment.
I agree because you should never claim "you'll be hearing from my lawyer". You should
However, a letter from a lawyer is not horribly expensive. If you draft a very precise letter in advance, and ask them to look over it at their normal rate, and correct any issues, you can get in and out in 1-2 hours. It would be likewise a couple of hours or less if you go to a lawyer that deals in trademark law regularly--he'll have a bunch of pre-fab letters and his secretary will pick the right one for your circumstances. Yes, that's $100-$400 (lawyer rates vary, and how much time they'll want to bill will vary--call around), but you're very likely to nip the problem in the bud and never hear from them again.
If they choose to continue to harass you, your lawyer (yes, now you really have a lawyer and when you wave around "you'll hear from my lawyer" they'll know you mean it), will likely take the case at reasonable rates or on contingency. If you have proof of using the mark for any period of time before they used it or claimed it, or if you are clearly in an unrelated field (or both), then the case is open and shut and few lawyers would turn it down.
The good news, however, is that no lawyer would come after you under such circumstances because the case is unwinnable, if they know you already have a lawyer and are not afraid to use it.
Responding to them personally is the worst thing you can do. Ignore them with prejudice, or have a lawyer send them a letter. Do not respond yourself unless you have training in trademark law, and even then you probably shouldn't respond yourself.
Stupid is so cool that we (meaning you all who voted for him) elected an intelligence underachiever (Bush) rather than a brainiac (Gore).
I agree with your theory, but your example is not a very good one. Gore is also an idiot, though distinctly less of an idiot than our current idiot-in-chief. I can't think of the last intellectually impressive president or presidential candidate we've had. Kennedy, certainly. Nixon, possibly (though paranoia has a negative impact on the clarity of ones thoughts), just to cover both sides of the right-left political spectrum. And though they both were seemingly quite smart, they still managed to do the wrong thing on so many counts it is embarassing, with Vietnam producing many of their biggest mistakes.
I believe America is so far down the path of deifying stupidity that we don't even get viable candidates that have half a brain, much less elect them. I'd like to think that there are lots of really smart people behind the curtains actually making the big decisions (like Greenspan in the Federal Reserve), but the behavior of our government pokes holes right through that theory.
It's disappointing, but par for the course, I guess. Large groups of people acting together are capable of stupidity on a massive scale. It shouldn't be surprising that they would elect a mindless jackass to lead the charge. It is sad to live in a time when the stupidest man to ever hold the office of President might just get reelected. At least the first time through, folks might have been able to claim ignorance...but this time around, everyone knows he's dumber than a box of rocks, and he's still polling in the lead.
And I kinda thought the internet and the dot-com boom would give nerds an edge in the political process...but I suppose we still actually have to get involved in the political process, which isn't nearly as much fun as hacking machines.
3) Go to eMusic.com, buy music in mp3 format
4) Go to AudioLunchbox.com, buy music in mp3 format
DRM-crippled formats aren't worth my time when my favorite records are available in mp3 from convenient, legal sources.
Mr. Shyamalan has a real gift for creating moods. He should consider moving to France and revitalizing the French film industry. No seriously. They can't afford the huge Hollywood budgets and have never recovered from the death of Francois Truffaut. He could become a great director like Erich Rohmer there instead of being forced to make stupid films due to Hollywood mass-appeal requirements.
No, seriously. Shyamalan is a jackass who makes really horrible movies (Sixth Sense being a surprising exception--I have to wonder if he had help from an uncredited smarter and more interesting friend).
French film, on the other hand, is actually quite healthy these days. Jean-Pierre Jeunet (City of Lost Children, Amelie), Krzysztof Kieslowski (Trois couleurs, though he died in '96 and didn't work exclusively in France, he's worth watching), François Ozon (Swimming Pool), and the list goes on. A lot of really great films are coming out of France, and overall I've found I enjoy a higher percentage of the French films I see than American films. Perhaps it is merely a filtering effect at work, and only the really exceptional French films see US releases.
It would be nice if Shyamalan would get filtered somewhere on the path to making his next film, so I don't even have to see the embarrassing previews.
I once read a tip about website passwords where you shouldn't have the same password for all sites that need a logic. One of the best suggestions I read was to have a password of say 4 characters, and intersperse the website name into it.
Ah, I get it. So that when the slashdot guys read your password 1s2l3a4s, they'll be able to guess that your yahoo mail password is 1y2a3h4o. Perfect! It's a system whereby you use a different password on every site, but still manage to make it available to anyone who has one of your passwords. In-genius. Really. I mean it.