I've been shopping for old computer crap on eBay lately (for nostalgia rather than collectibility), and I suspect your 1983 NIB Atari 5200 trackball would bring ten, maybe fifteen bucks (but I haven't been shopping for Atari game gear, so I'm really guessing). If ten or fifteen bucks, and reclaiming the space it takes up in your house, is worth more than the trackball to you, you should sell it. Part of the fun of these old machines and things is that they are dramatically cheaper than when we were kids. We couldn't have every cool peripheral and game back then, because it would have been cost prohibitive. Today, with stuff going for tens of dollars, even things that were very expensive back then, we can pick up just about anything we like and satisfy those old lingering curiosities. And, then, when we get bored with it...pass it on to someone else at about the same low price.
I just bought a Koala and a Commodore 64 on eBay this week. I've been feeling nostalgic of late, and started making chiptunes using VICE and GoatTracker (a SID composition tool for Linux and Windows), and got to thinking that I'd enjoy tinkering with the real thing. Saw the Koala going for like eight bucks on eBay and couldn't resist.
Sorry the term pisses you off so. I view it as nothing more than a convenient term to cover a set of concepts. I don't think we're in agreement on what it means, however, and I can see how you'd find it irritating to use a new term for something that seems like an old concept to you.
I promise I'll never force you to use the term "cloud computing". But, to say that it's the same as "remote computing" is to say, "I don't know what one of these two terms means". Cloud computing (or whatever you want to call it) brings along with it certain expectations that never existed with "remote computing" of old--things like ssh, VNC, remote desktop, X11, provide none of the qualities that make something a "cloud" service...except possibly, "available everywhere", by some very limited definitions of "available" and "everywhere".
"Right , so in other words its a catch all term for remote computing."
Does "remote computing" connote any of the qualities I mentioned? I never thought it did...and I've been using "remote computing" in the form of VNC, remote desktop, ssh, etc. for a decade or more. I never had an expectation any of those qualities applied to any of those services.
I didn't claim any single one of those qualities is new...the point is that the term (like the term AJAX) is a concise way of describing a set of qualities that previously existed.
And, if you believe resource expandability/shrinkability is only available in mainframes, I'd humbly suggest you look at Xen on Linux and Zones on Solaris. Note I didn't say this up/down sizing had to be infinite, or across multiple machines...and, of course, mainframes do not provide infinite sizing, either. It merely needs to scale up and down to fit the size of a given task. For the vast majority of computing tasks, a small piece of a single CPU will do.
You're putting words in my mouth and then disputing those words.
Data availability - You can get to it from anywhere. It's on the web, or it's accessible via some sort of network call using a standard protocol.
Data portability - You can move it from one location to another...including a local machine or another node in the "cloud". (Gmail provides IMAP access, for example...a standard protocol, allowing you to use your data in ways Google never thought of.)
Resource expandability/shrinkability - You can use whatever amount of space/CPU you need, and buy more on demand. When you don't need it anymore, you can give it back, quickly and easily...either completely automatically or via API calls.
A "cloud" service can be an end-user service, like Gmail, or it can be a developer-targeted service, like AWS.
Ideally, some standards will emerge, reliability will improve, and we'll all find "cloud computing" to be a reasonable part of our infrastructure, no matter what we do. And, those pieces of the puzzle that are currently missing from the existing "cloud" services will hopefully find there way into place.
So, the name is being abused...but the concepts that it concisely represents are good concepts. They're good for consumers and good for developers, and they will be a part of our lives going forward (and we'll like it!). If you're building web services and you're not trying to figure out how to provide availability, portability, and right-sizing of resources, then you're going to miss out on lots of opportunities to have less negative impact on the environment, and lots of opportunities to save your users time and money.
The Internet is not "the cloud". The Internet is the way cloud services are delivered. Without the basic characteristics I mentioned, you're just building regular old web applications.
"I'd venture a guess that there is some very advanced networking code in use at Google"
"networking code"? They use the stock Linux network stack just like the rest of us (actually a pretty old version of it on most of their machines), and I'm kind of baffled that you'd think that networking code is where Google would be doing interesting work. Networking is a solved problem. It's what you do with the network that interesting, and where Google is spending their money and time.
But, since you haven't bothered to actually, you know...look at their Open Source projects website, I'll mention that Google does have a reasonable stable of open source projects that they've released...a million lines or so of it, apparently. http://code.google.com/hosting/projects.html
They are heavy users of MySQL, and I know that they've paid for quite a bit of MySQL development work. And they employ too many OSS project developers to count.
Take your GPL defensiveness somewhere where it's useful. Google is a pretty good member of the OSS community...you wouldn't want us to start demanding that you start working on Open Source projects right away just because you use gThumb to organize your porn, would you?
He has hired a suit, just last week. Sheryl Sandberg, former VP of Global Online Sales and Operations at Google (the division that makes 99% of Google's money) and the 29th most powerful woman in the world according to Fortune magazine, will be sharing a cube with Zuckerberg, presumably to provide some adult supervision. She's under 40, but she's got plenty of experience...I've never met her, but I know a half dozen people that worked for her, and all think very highly of her.
It's perhaps the only hire I could imagine Facebook making that would give it a snowball's chance in hell of being worth the 15 billion valuation that the MS investment gave them last year. I still probably wouldn't consider Facebook options very valuable, but the odds of them being valuable definitely just went up.
One could have been working with Perl, UNIX, and Sendmail while cc:mail, ColdFusion, and NetWare were ruling the roost, and your skills would still be valid and valuable today. Actually, a lot of us probably were doing just that...
I dunno about you, but this tells me that relying on exclusively proprietary technologies is risky, at best. Not just from a business perspective, but from a personal development perspective, as well. So, IT guys are shooting themselves in the foot when they choose the latest new thing from Microsoft (Microsoft has built and killed numerous technologies during that same period...I'm surprised Visual Basic and FoxPro weren't on the list...they were bigger than several of the mentioned technologies and have fallen farther).
I guarantee you in 5 years your perspective will be different. (And my business's valuation will be an order of magnitude bigger... as will my percentage ownerhsip.)
Care to make it interesting? Your $15k "cash on hand" might be a good starting point for a wager.;-)
Actually, many investors in the valley wouldn't think very highly of your business acumen if you risked your house. Part of what investors look for is someone that is capable of recognizing low risk opportunities at low cost (the bigger the opportunity, the more risk and cost they're will to accept). Mortgaging ones house is a high cost and high risk action, with no solid upside. I'm not saying its the wrong way to fund a company, if you really have strong indications that it will pay off big, but I am saying that it is not something investors are going to look favorably on--in fact, I wouldn't mention it if you decide to talk to investors. Self-funding is fine, if you've got some money you can risk--I funded my first company with ~40 grand I made in the stock market, which isn't a problem. Then I occasionally funded it with credit card debt, when things got tight. In hindsight, I see that was a mistake...I should have been seeking outside funding before it got so tight. Things would have turned out better with that business had I done so (it fizzled out with moderate grace...having paid me a salary for seven years and bought me a nice car, I closed up shop at the end of 2005, and no one really noticed).
Most live in San Francisco sharing ~$2700 2 bedroom apartments. I rented a house in Mountain View for $2000/month, because I have a girlfriend and a big dog and they like to have a little room (and my co-founder already leases a town home in Santa Clara and has a family). My cut of our $15k would cover three months of that along with my food and other expenses (just barely). It's entirely possible for folks who live frugally to live in the valley for three months for $7500. I lived a bit rich, by choosing a house instead of an apartment.
Since you aren't going out drinking at night (you've got work to do), you're eating in mostly (you've got work to do), you aren't dating (you've got work to do), and you don't even need a car (you're centrally located, so you can share a Zip car with other founders or ride the train or walk to the weekly dinners), you don't have a lot of expenses. Mostly these are Web 2.0 companies, where the biggest expense is ~$100/month for a server.
Alright, there's a lot of misapprehension, and perhaps willful ignorance, of nearly every aspect of the Y Combinator model. I'm the co-founder of a company that was funded during the Winter Founders Program documented in the linked article (my company is Virtualmin, Inc., co-founded with Jamie Cameron).
First up, about the equity. YC asks for between 2% and 10%. Mostly, it's 5 or 6 percent. The companies funded are all quite early stage. It's very rare for one to be launched, and even moreso for it to be profitable. In some cases, it's no more than a mocked up demo. YC are very early stage investors, usually getting involved before anyone else will touch it--they invest in smart people, not really ideas or businesses. In the three months I've spent meeting once a week with the other founders, no one has ever even hinted that they regret giving up equity to YC. We certainly don't, and we're one of the few that had a launched product and paying customers.
The "5 minute IQ test" is being misconstrued. Applicants know well in advance exactly what the terms are going to look like. You don't bother applying if you don't like the terms, so you never get to the yes/no IQ test. I don't think anyone has ever turned them down at that stage. It's a good punchline, and makes for good magazine copy, nothing more.
Paul Graham is extremely smart. Wherever he goes crowds gather round, and it's not just for his boyish good looks. He's got a touch of ADD, at this point, due to his popularity, but we've never had trouble getting advice when we needed it, and his advice has generally been spot-on. YC has three other partners, two of whom (Jessica and Trevor) were as deeply involved as Paul during WFP. Paul's celebrity leads to a ridiculous array of speakers at the weekly dinners...He has an uncanny knack for bringing in the most interesting people in the valley: Joe Kraus (Excite, JotSpot), Evan Williams (Blogger, Twitter), Paul Buchheit (Gmail), Greg McAdoo (Sequoia), Ron Conway (largest angel investor in the world), etc.
Which brings me to contacts. If you believe contacts don't mean anything, you're fooling yourself. I started a business outside of the valley in 1999, and now I've started one in the valley. Big difference. I paid my bills and bought myself a nice car with my previous business. I have much higher expectations with my current business, and a large percentage of those expectations have been brought nearer by our affiliation with YC. Try dropping a random investor an email sometime, to arrange a meeting to tell them about your great business. We've never received a "no thanks" to such a meeting, and I've been hearing from fellow YC'ers that they've always gotten the meetings they wanted (not just random VCs...they're talking to exactly the people they want to talk to). We're in talks with our first choice VC and it's going very well, and at least three of the other companies have already closed rounds. If you are in Y Combinator you increase your chances of getting funded by a good investor by a huge amount (and let's be clear: A bad investor brings nothing but money and disaster will follow...the right investor brings more contacts, good advice, expertise in the right areas, and also money...with the right investor your odds of explosive success are remarkably higher). YC brings the best contacts in the industry.
Some other bits that aren't obvious unless you think it through and actually read the YC information on their page:
YC pays for the incorporation and all legal stuff for issuing shares. It's about ten grand worth of legal work from a top valley firm.
YC feeds the company founders every Tuesday night for three months. Paul cooks the meals personally (I've seen it with my own two eyes). These dinners are the single most valuable aspect of the program (aside from providing the motivation needed to get people out to the valley). Chatting with fellow founders every week about what you're working on, what they're working on, and exchanging ide
I don't do it, and I've never been in a situation where I felt like I needed to be able to do it. If it's a critical server, it has a hot spare disk (which is supported by Linux software RAID) so I can schedule downtime at the earliest convenient time. It it's not critical, I schedule downtime roughly immediately (i.e. end of business day, or whatever fits). I don't have any systems that must always run, forever and ever, amen. I just don't. I'm sure there are situations where it's necessary, but I've never seen one.
I used to believe in hardware RAID, but I've come into enough situations where the prior admin had setup critical systems with hardware RAID and were using whatever controller Dell shipped them--and they didn't have extras on-site. When the RAID controller failed, and I've seen them fail enough to consider it a real threat, the wait was hours or days while Dell got a new controller on-site. Sometimes at great expense.
Thankfully, I don't have to worry about that sort of thing these days (might have to again in the future, but IT admin is in my past for now). But at least a few of my most hair-raising IT experiences were due to problems with hardware RAID controllers. I'm not saying they don't do their job, or that a good RAID controller doesn't do a better job (for some definitions of "better") than software RAID. I'm saying that if you don't have a spare for every RAID controller in your office available, on-site, you're going to regret it one day. Even if you do, you might still regret it--as I mentioned, I've met two identical model Adaptec controllers that were incompatible due to different BIOS versions. Data was lost and had to be recovered from backups. I've also managed to screw up the RAID table due to human error (sometimes those Sunday late night upgrades go awry) in ways that would have been less likely to happen and more easily repaired in Linux, but the limited nature of the OS available on a hardware RAID controller made it far easier to screw up and more difficult and time consuming to repair. Having a whole OS with excellent partition and array management tools and a real shell to explore the situation with is a mighty fine thing when trying to figure out what stupid thing you did that led to a system not finding any system disks on boot.
If you need hot swap, you'll need a controller that supports it...though I feel obligated to mention that hot swap is a part of the SCSI and SATA standards, not any magical RAID feature. It can be implemented on any SCSI or SATA controller, regardless of RAID capability, and some Linux drivers probably support it (though I've never needed it, so I don't know for sure). Though it's probable that the controllers that most often have support for the hot-swap features are the ones also equipped with RAID capabilities.
Err, NO! It's about FAKERAID, which is a H/W S/W combo.
RAID stands for Redundant Array of Inexpensive/Independent Disks. Nowhere does it say "Controlled By A Dedicated CPU" ("RAIDCBADC"? Doesn't quite sing like "RAID"). Software RAID is as much RAID as a top of the line server RAID controller with RAM and a battery backup. It isn't as fast, sure, and it loads the system CPU, but it is still RAID. Calling it "FAKERAID" is just pretentious and misleading. The data integrity benefits are still present, as are some performance benefits in some circumstances (in fact, Linux RAID is demonstrably faster in some workloads than a top end Adaptec hardware RAID controller, though this is the exception rather than the rule)
That said, I hate pretty much all RAID controllers (whether software or hardware). Linux software RAID means that I can drop the disks into any PC and access the data. Every RAID controller from Promise, Adaptec, and Tektronic requires me to use their disk format, and if I lose the controller I lose the data until I can get another controller. Sure, in high availability environments, you keep a spare...but with Linux software RAID, every PC in the office is a spare controller. That's my kinda redundancy. I've even had two identical Adaptecs with different firmware lead to pretty massive data loss during a server migration. Thankfully there were good backups. I've never had similar problems moving Linux software RAID disks into a new Linux box.
One annoyance I had personally was that it didn't use Debians way of dealing with apache virtual hosts, and the config file ended up being one big unreadable mess as a result.
Fixed in both GPL and Professional, at least I'm pretty sure Debian's way is supported--it's all configurable options now. We added support for vhosts.conf, or something along those lines for SUSE...all virtual hosts go in a separate file. I consider that an unreadable mess, since I like it all in one place, but I can see how some folks might like it separate.;-)
If the One True Debian way is not supported today, it will be in a week or so when Virtualmin Professional gets support for Debian.
One advice i would give is that as a Debian user, I expect most everything I install to come in nice little.deb packages. I think the effort could be better spend on making debian packages, compared to creating a custom installer.
Advice appreciated, but not needed. All Virtualmin Professional packages are in native formats (RPM for Red Hat, Mandriva, SUSE, Fedora; and when I get them supported:.debs for Ubuntu and Debian, ports for FreeBSD, Portage for Gentoo). Webmin on the other hand is an Open Source project--we don't make any money on it except peripherally. No one has ever stepped up to help us maintain Debian packages (and the ones that were in Debian have long had problems with being unmaintained for long stretches, as you noted, and there was quite a strong antagonism from some Debian users towards Webmin itself, based primarily on the packaging rather than Webmin...no bugs filed in the Webmin bug tracker, but a lot of angry "Webmin sucks because it doesn't do things the Debian way out of the box" bugs in the Debian bug tracker that I only noticed months or years later when researching Virtualmin support for Debian...Webmin is extremely configurable, and often the Debian way is supported if you just configure it to do it that way, but if we haven't received config file updates to correct the defaults, we don't know we're doing it wrong...we can't test all 200+ modules on all 50-ish supported OS variants, or we'd never do anything else). We've long since fixed every issue I'm aware of that the Debian folks had problems with, including the one you've mentioned specifically, but no one is maintaining debs at this point. I will be doing so soon, however, and if the Debian powers that be will let me, I'll drop them into the public Debian repository. My biggest fear in that regard is that they'll want the package to be the way it was packaged previously (every module in a separate package)...I suspect this is a source of major problems of maintenance and dependency resolution. On RPM systems we provide one package for Webmin and one package for Usermin. Third party and additional modules and themes are provided separately in their own RPMs, but Webmin as a whole kind of expects to be whole--a lot of modules rely on others to do their job and breaking them all out into pieces is perfectly feasible, but getting the deps right in the packages is going to be very difficult and time consuming.
Which means it knows about how Debian does things and doesn't screw up your system, like virtualmin for instance.
Have you filed a bug? We're unaware of any issues on Debian at this time--there were some issues with rc.d entries about a year ago, but they've long been resolved. I'm not sure what else you're having trouble with. File a bug about it, and we'll get it fixed. I've been working on a version of the Virtualmin Professional installer for Debian for the past couple of weeks, and things are going reasonably well (and as far as I know, Virtualmin GPL and Webmin work fine on Debian). Anyway, let us know what goes wrong for you with Virtualmin on Debian here:
Virtualmin Professional also provides an application installation feature, wherein you can have Wordpress, Django, etc. installed for you. Rails, and a few other doodads, will show up in a near future release...installing software automatically isn't a difficult task, merely time-consuming to write the scripts and maintain them. We currently support 51 apps in Virtualmin, with more being added as users ask for them.
I'm not sure what bugs you about Webmin, other than the ugly default themes. Complexity is minimized by not doing things behind your back--it is a one to one mapping of configuration directives to GUI options, with a few helpers here and there to make sure you get all the right bits into the right places. It respects comments and configuration file order, it doesn't modify options it doesn't understand, it is compatible with doing things the system standard way on all popular platforms (e.g. it starts and stops services with "/etc/init.d/servicename start|stop"), and you can use a text editor to modify any configuration that Webmin makes. The same cannot be said for any other general system administration tool available, that I am aware of. Not to mention, Webmin simply has no competition in the general system administration interface market since there is no product that administers anywhere near as many services. But we're talking about virtual host administration tools, which are much more limited in scope but more ambitious in depth than Webmin, so Webmin shouldn't really be in the discussion, except to say that it makes a great base upon which to build a virtual host administration system. I just think you're doing folks a great disservice by directing them away from Webmin--regardless of what you think of it for virtual host administration (which would be an uncomfortable fit) as a general system administration tool it is unrivalled, and it is free. (Oh, and six+ million downloads gives me some confidence that not everyone agrees with your assessment.)
Anyway, to each his own, but I reckoned I'd chime in with a conflicting view (I am one of the developers of Virtualmin GPL and Professional and a long-time documenter and cheerleader for Webmin, so take it with a grain of salt, if you like). The nice thing is that Webmin and Virtualmin GPL are free, and Virtualmin Professional has an online demo. Try them and decide for yourself.
They're similarly gutting Wikibooks- Jimbo just came in one day and said anything that isn't a textbook for a college or high school course had to go. All the guides, how-tos, and anything that wasn't in the strictest sense academic. Both resources are being horribly mismanaged by the wikimedia board.
When I visited wikibooks for the first and last time, I was completely turned off by the absolutely laughable and bizarre topics of the books (and the fact that most of them seemed to have been started as wishes for someone to write the book rather than intention to write significant portions...they were all absolutely devoid of content and hadn't been edited in weeks or months...kinda like what I would expect if anyone can say "I'm gonna write a book!"). By your description it sounds like the changes are just the kind of thing that might make me consider it worth visiting again.
HiThere is right about taking a job with shifty companies, but there are a lot of other reasons to simply avoid this kind of deal.
I've had occasion to deal with good folks and bad in running my own business, sometimes taking on projects I didn't really believe in (not particularly evil or destructive, just not in a direction that I felt was worth pursuing and thus a waste of my time) in order to pay the bills. These projects often promised far bigger returns than more fun, interesting, and good projects. However, in almost every case, I wound up spending more time and money on the project than anticipated and getting less money in the end. The companies footing the bills reneged on promises, disappeared into insolvency, redirected their efforts midway through the project (while I'd turned down other jobs in anticipation of this one being extremely long-term and high demand), etc. Sure, a strong contract (which is often where the first hints of trouble come up--getting the contract signed before work begins is often like pulling teeth with this kind of company...everything is always upbeat and enthusiastic, but discussions are long and fruitless). This is all very vague sounding, I know, but in seven years of running my first business I learned something very important:
The fact is, if you're doing what you believe in, you're far more likely to make a sustainable living than if you're taking on big money but shifty projects or jobs.
"Shifty" is the vague bit here. Seemingly large and hugely successful dotcom boom companies have often been the ones that wasted the most of my time and effort and made me feel far less satisfied with my work. On the other hand, one of the more pleasant work experiences was for an internet pornography company. They're probably the largest such company in the world now, but it was relatively small back then. I could tell immediately that the guy running the show was technically savvy, understood his needs very well, and had the authority to sign off on the work and pay for it; and I could see that our work could vastly improve the performance of his website, and thus his customer satisfaction, at a good price. The point is that the good projects and jobs are straightforward, clearly defined, technically sound and interesting, and payment is well-defined (and usually fair--not too high, not too low). Contests, commission-based pay, projects that you can see are clearly foolish and won't generate profit for your client, projects to create one-off software that already has a large market leader in the field with thousands of users, etc. are all warning signs of a bad deal for you no matter how good the payment looks. Further, those who are most ready to give you everything you ask for in initial negotiations are also probably the most difficult to deal with (and least likely to pay you enough to cover the time spent on them, and most likely to cause trouble during the actual contract signing).
What I'm trying to get at, is that in the real world, you can probably make a living working for people like The SCO Group. But you probably won't enjoy it, you'll probably get paid less than you deserve for the soul-sucking work, and you'll probably lock yourself into doing jobs for companies just like The SCO Group in the future. Not just because you have The SCO Group on your resume, and good tech companies will look down on you for it (which many will), but because your experiences will be in making oddball poorly designed products for a shifty company--the situation will demand that that's what happens with your project, even if you're a great developer. Working for a good company with a good vision leads to good products and your subsequent job offers will get better over time rather than worse. Go for good companies that do things you can believe in, even if you have to start out making less than the shifty company is offering. Working on something great is far more valuable to your longterm economic success than being paid an exhorbitant wage for working
Slashdot Mirror
Probably due to repeated hammer blows (delivered by daddy) to their heads as they grew up.
Don't you mean "delivered by Jews"? Jews are always hitting children over the head with hammers, don't you know.
I've been shopping for old computer crap on eBay lately (for nostalgia rather than collectibility), and I suspect your 1983 NIB Atari 5200 trackball would bring ten, maybe fifteen bucks (but I haven't been shopping for Atari game gear, so I'm really guessing). If ten or fifteen bucks, and reclaiming the space it takes up in your house, is worth more than the trackball to you, you should sell it. Part of the fun of these old machines and things is that they are dramatically cheaper than when we were kids. We couldn't have every cool peripheral and game back then, because it would have been cost prohibitive. Today, with stuff going for tens of dollars, even things that were very expensive back then, we can pick up just about anything we like and satisfy those old lingering curiosities. And, then, when we get bored with it...pass it on to someone else at about the same low price.
I just bought a Koala and a Commodore 64 on eBay this week. I've been feeling nostalgic of late, and started making chiptunes using VICE and GoatTracker (a SID composition tool for Linux and Windows), and got to thinking that I'd enjoy tinkering with the real thing. Saw the Koala going for like eight bucks on eBay and couldn't resist.
Sorry the term pisses you off so. I view it as nothing more than a convenient term to cover a set of concepts. I don't think we're in agreement on what it means, however, and I can see how you'd find it irritating to use a new term for something that seems like an old concept to you.
I promise I'll never force you to use the term "cloud computing". But, to say that it's the same as "remote computing" is to say, "I don't know what one of these two terms means". Cloud computing (or whatever you want to call it) brings along with it certain expectations that never existed with "remote computing" of old--things like ssh, VNC, remote desktop, X11, provide none of the qualities that make something a "cloud" service...except possibly, "available everywhere", by some very limited definitions of "available" and "everywhere".
Oh, yeah, thanks for the heads up about the error.
"Right , so in other words its a catch all term for remote computing."
Does "remote computing" connote any of the qualities I mentioned? I never thought it did...and I've been using "remote computing" in the form of VNC, remote desktop, ssh, etc. for a decade or more. I never had an expectation any of those qualities applied to any of those services.
I didn't claim any single one of those qualities is new...the point is that the term (like the term AJAX) is a concise way of describing a set of qualities that previously existed.
And, if you believe resource expandability/shrinkability is only available in mainframes, I'd humbly suggest you look at Xen on Linux and Zones on Solaris. Note I didn't say this up/down sizing had to be infinite, or across multiple machines...and, of course, mainframes do not provide infinite sizing, either. It merely needs to scale up and down to fit the size of a given task. For the vast majority of computing tasks, a small piece of a single CPU will do.
You're putting words in my mouth and then disputing those words.
It's my sig. Since when is an off-topic signature a problem?
It actually has a reasonable definition.
The point of "the cloud" is:
Data availability - You can get to it from anywhere. It's on the web, or it's accessible via some sort of network call using a standard protocol.
Data portability - You can move it from one location to another...including a local machine or another node in the "cloud". (Gmail provides IMAP access, for example...a standard protocol, allowing you to use your data in ways Google never thought of.)
Resource expandability/shrinkability - You can use whatever amount of space/CPU you need, and buy more on demand. When you don't need it anymore, you can give it back, quickly and easily...either completely automatically or via API calls.
A "cloud" service can be an end-user service, like Gmail, or it can be a developer-targeted service, like AWS.
Ideally, some standards will emerge, reliability will improve, and we'll all find "cloud computing" to be a reasonable part of our infrastructure, no matter what we do. And, those pieces of the puzzle that are currently missing from the existing "cloud" services will hopefully find there way into place.
So, the name is being abused...but the concepts that it concisely represents are good concepts. They're good for consumers and good for developers, and they will be a part of our lives going forward (and we'll like it!). If you're building web services and you're not trying to figure out how to provide availability, portability, and right-sizing of resources, then you're going to miss out on lots of opportunities to have less negative impact on the environment, and lots of opportunities to save your users time and money.
The Internet is not "the cloud". The Internet is the way cloud services are delivered. Without the basic characteristics I mentioned, you're just building regular old web applications.
"I'd venture a guess that there is some very advanced networking code in use at Google"
"networking code"? They use the stock Linux network stack just like the rest of us (actually a pretty old version of it on most of their machines), and I'm kind of baffled that you'd think that networking code is where Google would be doing interesting work. Networking is a solved problem. It's what you do with the network that interesting, and where Google is spending their money and time.
But, since you haven't bothered to actually, you know...look at their Open Source projects website, I'll mention that Google does have a reasonable stable of open source projects that they've released...a million lines or so of it, apparently. http://code.google.com/hosting/projects.html
They are heavy users of MySQL, and I know that they've paid for quite a bit of MySQL development work. And they employ too many OSS project developers to count.
Take your GPL defensiveness somewhere where it's useful. Google is a pretty good member of the OSS community...you wouldn't want us to start demanding that you start working on Open Source projects right away just because you use gThumb to organize your porn, would you?
He has hired a suit, just last week. Sheryl Sandberg, former VP of Global Online Sales and Operations at Google (the division that makes 99% of Google's money) and the 29th most powerful woman in the world according to Fortune magazine, will be sharing a cube with Zuckerberg, presumably to provide some adult supervision. She's under 40, but she's got plenty of experience...I've never met her, but I know a half dozen people that worked for her, and all think very highly of her.
It's perhaps the only hire I could imagine Facebook making that would give it a snowball's chance in hell of being worth the 15 billion valuation that the MS investment gave them last year. I still probably wouldn't consider Facebook options very valuable, but the odds of them being valuable definitely just went up.
Those bastards owe me about 300 bucks! I finally gave up on rebates. In fact, now if it's got a rebate offer on it, I buy a competing item.
One could have been working with Perl, UNIX, and Sendmail while cc:mail, ColdFusion, and NetWare were ruling the roost, and your skills would still be valid and valuable today. Actually, a lot of us probably were doing just that...
I dunno about you, but this tells me that relying on exclusively proprietary technologies is risky, at best. Not just from a business perspective, but from a personal development perspective, as well. So, IT guys are shooting themselves in the foot when they choose the latest new thing from Microsoft (Microsoft has built and killed numerous technologies during that same period...I'm surprised Visual Basic and FoxPro weren't on the list...they were bigger than several of the mentioned technologies and have fallen farther).
Care to make it interesting? Your $15k "cash on hand" might be a good starting point for a wager. ;-)
Actually, many investors in the valley wouldn't think very highly of your business acumen if you risked your house. Part of what investors look for is someone that is capable of recognizing low risk opportunities at low cost (the bigger the opportunity, the more risk and cost they're will to accept). Mortgaging ones house is a high cost and high risk action, with no solid upside. I'm not saying its the wrong way to fund a company, if you really have strong indications that it will pay off big, but I am saying that it is not something investors are going to look favorably on--in fact, I wouldn't mention it if you decide to talk to investors. Self-funding is fine, if you've got some money you can risk--I funded my first company with ~40 grand I made in the stock market, which isn't a problem. Then I occasionally funded it with credit card debt, when things got tight. In hindsight, I see that was a mistake...I should have been seeking outside funding before it got so tight. Things would have turned out better with that business had I done so (it fizzled out with moderate grace...having paid me a salary for seven years and bought me a nice car, I closed up shop at the end of 2005, and no one really noticed).
Most live in San Francisco sharing ~$2700 2 bedroom apartments. I rented a house in Mountain View for $2000/month, because I have a girlfriend and a big dog and they like to have a little room (and my co-founder already leases a town home in Santa Clara and has a family). My cut of our $15k would cover three months of that along with my food and other expenses (just barely). It's entirely possible for folks who live frugally to live in the valley for three months for $7500. I lived a bit rich, by choosing a house instead of an apartment.
Since you aren't going out drinking at night (you've got work to do), you're eating in mostly (you've got work to do), you aren't dating (you've got work to do), and you don't even need a car (you're centrally located, so you can share a Zip car with other founders or ride the train or walk to the weekly dinners), you don't have a lot of expenses. Mostly these are Web 2.0 companies, where the biggest expense is ~$100/month for a server.
Why, yes, I did. Thanks for asking.
Alright, there's a lot of misapprehension, and perhaps willful ignorance, of nearly every aspect of the Y Combinator model. I'm the co-founder of a company that was funded during the Winter Founders Program documented in the linked article (my company is Virtualmin, Inc., co-founded with Jamie Cameron).
First up, about the equity. YC asks for between 2% and 10%. Mostly, it's 5 or 6 percent. The companies funded are all quite early stage. It's very rare for one to be launched, and even moreso for it to be profitable. In some cases, it's no more than a mocked up demo. YC are very early stage investors, usually getting involved before anyone else will touch it--they invest in smart people, not really ideas or businesses. In the three months I've spent meeting once a week with the other founders, no one has ever even hinted that they regret giving up equity to YC. We certainly don't, and we're one of the few that had a launched product and paying customers.
The "5 minute IQ test" is being misconstrued. Applicants know well in advance exactly what the terms are going to look like. You don't bother applying if you don't like the terms, so you never get to the yes/no IQ test. I don't think anyone has ever turned them down at that stage. It's a good punchline, and makes for good magazine copy, nothing more.
Paul Graham is extremely smart. Wherever he goes crowds gather round, and it's not just for his boyish good looks. He's got a touch of ADD, at this point, due to his popularity, but we've never had trouble getting advice when we needed it, and his advice has generally been spot-on. YC has three other partners, two of whom (Jessica and Trevor) were as deeply involved as Paul during WFP. Paul's celebrity leads to a ridiculous array of speakers at the weekly dinners...He has an uncanny knack for bringing in the most interesting people in the valley: Joe Kraus (Excite, JotSpot), Evan Williams (Blogger, Twitter), Paul Buchheit (Gmail), Greg McAdoo (Sequoia), Ron Conway (largest angel investor in the world), etc.
Which brings me to contacts. If you believe contacts don't mean anything, you're fooling yourself. I started a business outside of the valley in 1999, and now I've started one in the valley. Big difference. I paid my bills and bought myself a nice car with my previous business. I have much higher expectations with my current business, and a large percentage of those expectations have been brought nearer by our affiliation with YC. Try dropping a random investor an email sometime, to arrange a meeting to tell them about your great business. We've never received a "no thanks" to such a meeting, and I've been hearing from fellow YC'ers that they've always gotten the meetings they wanted (not just random VCs...they're talking to exactly the people they want to talk to). We're in talks with our first choice VC and it's going very well, and at least three of the other companies have already closed rounds. If you are in Y Combinator you increase your chances of getting funded by a good investor by a huge amount (and let's be clear: A bad investor brings nothing but money and disaster will follow...the right investor brings more contacts, good advice, expertise in the right areas, and also money...with the right investor your odds of explosive success are remarkably higher). YC brings the best contacts in the industry.
Some other bits that aren't obvious unless you think it through and actually read the YC information on their page:
YC pays for the incorporation and all legal stuff for issuing shares. It's about ten grand worth of legal work from a top valley firm.
YC feeds the company founders every Tuesday night for three months. Paul cooks the meals personally (I've seen it with my own two eyes). These dinners are the single most valuable aspect of the program (aside from providing the motivation needed to get people out to the valley). Chatting with fellow founders every week about what you're working on, what they're working on, and exchanging ide
What about hotswapping?
I don't do it, and I've never been in a situation where I felt like I needed to be able to do it. If it's a critical server, it has a hot spare disk (which is supported by Linux software RAID) so I can schedule downtime at the earliest convenient time. It it's not critical, I schedule downtime roughly immediately (i.e. end of business day, or whatever fits). I don't have any systems that must always run, forever and ever, amen. I just don't. I'm sure there are situations where it's necessary, but I've never seen one.
I used to believe in hardware RAID, but I've come into enough situations where the prior admin had setup critical systems with hardware RAID and were using whatever controller Dell shipped them--and they didn't have extras on-site. When the RAID controller failed, and I've seen them fail enough to consider it a real threat, the wait was hours or days while Dell got a new controller on-site. Sometimes at great expense.
Thankfully, I don't have to worry about that sort of thing these days (might have to again in the future, but IT admin is in my past for now). But at least a few of my most hair-raising IT experiences were due to problems with hardware RAID controllers. I'm not saying they don't do their job, or that a good RAID controller doesn't do a better job (for some definitions of "better") than software RAID. I'm saying that if you don't have a spare for every RAID controller in your office available, on-site, you're going to regret it one day. Even if you do, you might still regret it--as I mentioned, I've met two identical model Adaptec controllers that were incompatible due to different BIOS versions. Data was lost and had to be recovered from backups. I've also managed to screw up the RAID table due to human error (sometimes those Sunday late night upgrades go awry) in ways that would have been less likely to happen and more easily repaired in Linux, but the limited nature of the OS available on a hardware RAID controller made it far easier to screw up and more difficult and time consuming to repair. Having a whole OS with excellent partition and array management tools and a real shell to explore the situation with is a mighty fine thing when trying to figure out what stupid thing you did that led to a system not finding any system disks on boot.
If you need hot swap, you'll need a controller that supports it...though I feel obligated to mention that hot swap is a part of the SCSI and SATA standards, not any magical RAID feature. It can be implemented on any SCSI or SATA controller, regardless of RAID capability, and some Linux drivers probably support it (though I've never needed it, so I don't know for sure). Though it's probable that the controllers that most often have support for the hot-swap features are the ones also equipped with RAID capabilities.
Err, NO! It's about FAKERAID, which is a H/W S/W combo.
RAID stands for Redundant Array of Inexpensive/Independent Disks. Nowhere does it say "Controlled By A Dedicated CPU" ("RAIDCBADC"? Doesn't quite sing like "RAID"). Software RAID is as much RAID as a top of the line server RAID controller with RAM and a battery backup. It isn't as fast, sure, and it loads the system CPU, but it is still RAID. Calling it "FAKERAID" is just pretentious and misleading. The data integrity benefits are still present, as are some performance benefits in some circumstances (in fact, Linux RAID is demonstrably faster in some workloads than a top end Adaptec hardware RAID controller, though this is the exception rather than the rule)
That said, I hate pretty much all RAID controllers (whether software or hardware). Linux software RAID means that I can drop the disks into any PC and access the data. Every RAID controller from Promise, Adaptec, and Tektronic requires me to use their disk format, and if I lose the controller I lose the data until I can get another controller. Sure, in high availability environments, you keep a spare...but with Linux software RAID, every PC in the office is a spare controller. That's my kinda redundancy. I've even had two identical Adaptecs with different firmware lead to pretty massive data loss during a server migration. Thankfully there were good backups. I've never had similar problems moving Linux software RAID disks into a new Linux box.
One annoyance I had personally was that it didn't use Debians way of dealing with apache virtual hosts, and the config file ended up being one big unreadable mess as a result.
;-)
.deb packages. I think the effort could be better spend on making debian packages, compared to creating a custom installer.
.debs for Ubuntu and Debian, ports for FreeBSD, Portage for Gentoo). Webmin on the other hand is an Open Source project--we don't make any money on it except peripherally. No one has ever stepped up to help us maintain Debian packages (and the ones that were in Debian have long had problems with being unmaintained for long stretches, as you noted, and there was quite a strong antagonism from some Debian users towards Webmin itself, based primarily on the packaging rather than Webmin...no bugs filed in the Webmin bug tracker, but a lot of angry "Webmin sucks because it doesn't do things the Debian way out of the box" bugs in the Debian bug tracker that I only noticed months or years later when researching Virtualmin support for Debian...Webmin is extremely configurable, and often the Debian way is supported if you just configure it to do it that way, but if we haven't received config file updates to correct the defaults, we don't know we're doing it wrong...we can't test all 200+ modules on all 50-ish supported OS variants, or we'd never do anything else). We've long since fixed every issue I'm aware of that the Debian folks had problems with, including the one you've mentioned specifically, but no one is maintaining debs at this point. I will be doing so soon, however, and if the Debian powers that be will let me, I'll drop them into the public Debian repository. My biggest fear in that regard is that they'll want the package to be the way it was packaged previously (every module in a separate package)...I suspect this is a source of major problems of maintenance and dependency resolution. On RPM systems we provide one package for Webmin and one package for Usermin. Third party and additional modules and themes are provided separately in their own RPMs, but Webmin as a whole kind of expects to be whole--a lot of modules rely on others to do their job and breaking them all out into pieces is perfectly feasible, but getting the deps right in the packages is going to be very difficult and time consuming.
Fixed in both GPL and Professional, at least I'm pretty sure Debian's way is supported--it's all configurable options now. We added support for vhosts.conf, or something along those lines for SUSE...all virtual hosts go in a separate file. I consider that an unreadable mess, since I like it all in one place, but I can see how some folks might like it separate.
If the One True Debian way is not supported today, it will be in a week or so when Virtualmin Professional gets support for Debian.
One advice i would give is that as a Debian user, I expect most everything I install to come in nice little
Advice appreciated, but not needed. All Virtualmin Professional packages are in native formats (RPM for Red Hat, Mandriva, SUSE, Fedora; and when I get them supported:
Which means it knows about how Debian does things and doesn't screw up your system, like virtualmin for instance.
Have you filed a bug? We're unaware of any issues on Debian at this time--there were some issues with rc.d entries about a year ago, but they've long been resolved. I'm not sure what else you're having trouble with. File a bug about it, and we'll get it fixed. I've been working on a version of the Virtualmin Professional installer for Debian for the past couple of weeks, and things are going reasonably well (and as far as I know, Virtualmin GPL and Webmin work fine on Debian). Anyway, let us know what goes wrong for you with Virtualmin on Debian here:
http://www.virtualmin.com/bug-tracker
You don't have to be a customer and you don't have to be using Virtualmin Professional. Bugs get fixed, regardless of the version in question.
Webmin, CPanel, Ensim, they are all deep, complex, and ugly. Webmin is the worst.
I would humbly disagree about Webmin (really, to compare to cPanel and Ensim, you need to be talking about Virtualmin).
http://www.virtualmin.com/videos/tour
http://www.virtualmin.com/demo
Virtualmin Professional also provides an application installation feature, wherein you can have Wordpress, Django, etc. installed for you. Rails, and a few other doodads, will show up in a near future release...installing software automatically isn't a difficult task, merely time-consuming to write the scripts and maintain them. We currently support 51 apps in Virtualmin, with more being added as users ask for them.
I'm not sure what bugs you about Webmin, other than the ugly default themes. Complexity is minimized by not doing things behind your back--it is a one to one mapping of configuration directives to GUI options, with a few helpers here and there to make sure you get all the right bits into the right places. It respects comments and configuration file order, it doesn't modify options it doesn't understand, it is compatible with doing things the system standard way on all popular platforms (e.g. it starts and stops services with "/etc/init.d/servicename start|stop"), and you can use a text editor to modify any configuration that Webmin makes. The same cannot be said for any other general system administration tool available, that I am aware of. Not to mention, Webmin simply has no competition in the general system administration interface market since there is no product that administers anywhere near as many services. But we're talking about virtual host administration tools, which are much more limited in scope but more ambitious in depth than Webmin, so Webmin shouldn't really be in the discussion, except to say that it makes a great base upon which to build a virtual host administration system. I just think you're doing folks a great disservice by directing them away from Webmin--regardless of what you think of it for virtual host administration (which would be an uncomfortable fit) as a general system administration tool it is unrivalled, and it is free. (Oh, and six+ million downloads gives me some confidence that not everyone agrees with your assessment.)
Anyway, to each his own, but I reckoned I'd chime in with a conflicting view (I am one of the developers of Virtualmin GPL and Professional and a long-time documenter and cheerleader for Webmin, so take it with a grain of salt, if you like). The nice thing is that Webmin and Virtualmin GPL are free, and Virtualmin Professional has an online demo. Try them and decide for yourself.
They're similarly gutting Wikibooks- Jimbo just came in one day and said anything that isn't a textbook for a college or high school course had to go. All the guides, how-tos, and anything that wasn't in the strictest sense academic. Both resources are being horribly mismanaged by the wikimedia board.
When I visited wikibooks for the first and last time, I was completely turned off by the absolutely laughable and bizarre topics of the books (and the fact that most of them seemed to have been started as wishes for someone to write the book rather than intention to write significant portions...they were all absolutely devoid of content and hadn't been edited in weeks or months...kinda like what I would expect if anyone can say "I'm gonna write a book!"). By your description it sounds like the changes are just the kind of thing that might make me consider it worth visiting again.
HiThere is right about taking a job with shifty companies, but there are a lot of other reasons to simply avoid this kind of deal.
I've had occasion to deal with good folks and bad in running my own business, sometimes taking on projects I didn't really believe in (not particularly evil or destructive, just not in a direction that I felt was worth pursuing and thus a waste of my time) in order to pay the bills. These projects often promised far bigger returns than more fun, interesting, and good projects. However, in almost every case, I wound up spending more time and money on the project than anticipated and getting less money in the end. The companies footing the bills reneged on promises, disappeared into insolvency, redirected their efforts midway through the project (while I'd turned down other jobs in anticipation of this one being extremely long-term and high demand), etc. Sure, a strong contract (which is often where the first hints of trouble come up--getting the contract signed before work begins is often like pulling teeth with this kind of company...everything is always upbeat and enthusiastic, but discussions are long and fruitless). This is all very vague sounding, I know, but in seven years of running my first business I learned something very important:
The fact is, if you're doing what you believe in, you're far more likely to make a sustainable living than if you're taking on big money but shifty projects or jobs.
"Shifty" is the vague bit here. Seemingly large and hugely successful dotcom boom companies have often been the ones that wasted the most of my time and effort and made me feel far less satisfied with my work. On the other hand, one of the more pleasant work experiences was for an internet pornography company. They're probably the largest such company in the world now, but it was relatively small back then. I could tell immediately that the guy running the show was technically savvy, understood his needs very well, and had the authority to sign off on the work and pay for it; and I could see that our work could vastly improve the performance of his website, and thus his customer satisfaction, at a good price. The point is that the good projects and jobs are straightforward, clearly defined, technically sound and interesting, and payment is well-defined (and usually fair--not too high, not too low). Contests, commission-based pay, projects that you can see are clearly foolish and won't generate profit for your client, projects to create one-off software that already has a large market leader in the field with thousands of users, etc. are all warning signs of a bad deal for you no matter how good the payment looks. Further, those who are most ready to give you everything you ask for in initial negotiations are also probably the most difficult to deal with (and least likely to pay you enough to cover the time spent on them, and most likely to cause trouble during the actual contract signing).
What I'm trying to get at, is that in the real world, you can probably make a living working for people like The SCO Group. But you probably won't enjoy it, you'll probably get paid less than you deserve for the soul-sucking work, and you'll probably lock yourself into doing jobs for companies just like The SCO Group in the future. Not just because you have The SCO Group on your resume, and good tech companies will look down on you for it (which many will), but because your experiences will be in making oddball poorly designed products for a shifty company--the situation will demand that that's what happens with your project, even if you're a great developer. Working for a good company with a good vision leads to good products and your subsequent job offers will get better over time rather than worse. Go for good companies that do things you can believe in, even if you have to start out making less than the shifty company is offering. Working on something great is far more valuable to your longterm economic success than being paid an exhorbitant wage for working