Especially if you have other Macs in your office, you can leave OS X on it and have a nice little small office server. You could also throw Debian or Ubuntu on it and use it as you see fit.
The small form factor would make it easy for a developer to keep one on the (literal) desktop alongside a workstation. Personally, I'd use virtualization instead, but others may prefer having a physical box to play with.
I was going to suggest using virtualized environments as well, but that still leaves the admin with the task of automating the management of all his difference systems. Frankly, I've always approached this the way he's already doing it: a set of scripts that manage things.
I'm not aware of any systems that do all this for you while still being flexible enough to accommodate lots of unique requirements. The script-based systems I've employed over the years all followed some basic rules:
Increment the version number and make a new DNS entry for each committed changeset.
Only allow migrations to "test" from "dev", and from "test" to "prod".
Allow automatic reverting to a previous version in test or prod, but require manual merging of changesets from later revisions to put them back into the older upper stage.
I've successfully managed hundreds of sites and web apps in this manner, with minimal fuss. Virtualization adds in extra complexity in some respects, but makes other things easier as each VPS can have its own customized environment. As long as everything in dev, test, and prod uses the same base VPS environment, problems should be minimal.
Okay, so how much do you figure it will cost to outfit a server to be capable of supporting an arbitrary number of users, running extremely IO intestive reports on a shared volume whenever they feel like it, including the network infrastructure required to support this? Oh, and don't forget redundancy for the server.
Trust me, I've learned from experience that the local disk space works out to be much cheaper for this sort of thing.
That's a completely difference scenario, and you know it. How about addressing the actual use case in question? He's talking about hundreds of gigabytes, not tens of terabytes. By the way, many organizations would have the "20 TB file server" so everyone can access the data, not necessarily exclusively via the share.
You also didn't specify how often the bulk of your data changes (I'm guessing the bulk of his share doesn't change every day, and yours probably doesn't either). Add in the fact that a 1 TB drive costs about $90 (you can do better if you shop around) and your objection to simply doing incremental updates for a few hundred gigs of data to a local mirror looks pretty ridiculous. It gets even more absurd when you consider how fast you can rsync a crapload of data over even a 100 Mbit connection, let along increasingly prevalent gigabit networks.
Why slam the server with complicated indexing schemes, coupled with multiple users competing for all the data on a potentially frequent basis? That sounds like a much bigger headache than just taking the simple route, unless business requirements specifically stipulate a minute-by-minute ability to run reports on the data in question. Given that the submitter is dealing with very time windows for processing reports in the first place, I don't get the impression that this is the case.
I don't think you understand how rsync works. After the initial mirror, it only transfers changed files, which tends to be a really quick operation. For most organizations, that's going to be a hell of a lot less than the total disk usage on the shared volume.
Whether it defeats the purpose or not depends entirely on the organization's needs. If querying data every few hours in a local app is the objective, that can be met quite effectively with mirroring. Disk space is cheap.
“All our political forms are exhausted and practically nonexistent. Our parliamentary and electoral system and our political parties are just as futile as dictatorships are intolerable. Nothing is left. And this nothing is increasingly aggressive, totalitarian, and omnipresent. Our experience today is the strange one of empty political institutions in which no one has any confidence any more, of a system of government which functions only in the interests of a political class, and at the same time of the almost infinite growth of power, authority, and social control which makes any one of our democracies a more authoritarian mechanism than the Napoleonic state.”
- Jacques Ellul
Price controls are being enforced to prevent what the Finnish government might perceive as "price gouging." This means that no matter how ridiculously remote your location might be, you now have a legal "right" to broadband Internet access at some ISP's hefty expense. That's absurd.
That's not the point at all. The entire point of this story is that the Finnish government has guaranteed access to broadband Internet service to every single citizen in the nation.
I read the article in its entirety; it's ridiculously lacking in details. Unfortunately, so is every other "source" on this topic.
We have two possibilities here: (1) Finland is enforcing price controls (as would seem to be the case to prevent perceived "gouging" by ISPs who now must provide broadband service to anyone, regardless of how remote their location may be), or (2) Finland's government (actually their citizens via taxes, which are very high there) is footing the bill.
That's effectively what Finland's government is doing. They're stipulating that consumers have a legal right to broadband Internet access at a "reasonable" price.
... but seriously, how is access to a broadband Internet connection a legal right? Somebody please explain this to me, because the article doesn't give any supporting logic.
I need air to breathe, food to eat, clothes to wear, and a place to sleep at night. As much as I enjoy working in I.T. for a living, I do not need Internet access to survive.
Redirection can be accomplished by any number of means, one of which is simply telling a mail client to BCC an email address for all outgoing mail. File corruption happens all the time, and doesn't necessarily mean the accused had anything to do with it. In fact, outright mass file deletion would be more suspect in my book.
Of course, the odds are extremely good that nobody on Slashdot actually knows the full story, but the "evidence" as presented is absurdly weak for a hacking accusation.
Flyersrights.org, a nonprofit organization founded in 2007, had been investigating surface delays in air travel. According to the suit, Hanni exchanged information with Frederick J. Foreman, who worked for Metron Aviation, which was hired by the U.S. Federal Aviation Administration to study surface delays. The suit says Foreman provided information to Hanni with permission from Metron, including a report that fingered Delta as having excessive surface delays. Metron is also named in the suit.
During the correspondence, AOL informed Hanni that her e-mails, spreadsheets and lists of donors were being redirected to an unknown destination. Also, files on Hanni's computer became corrupted, the suit says. The hacking began in 2008 and continued through this year.
This does not constitute "hacking" (or even cracking, as it should be termed). Unless I've missed something here, the actual allegation is that information was improperly disclosed, but not that an email account was broken into.
Slashdot Mirror
Especially if you have other Macs in your office, you can leave OS X on it and have a nice little small office server. You could also throw Debian or Ubuntu on it and use it as you see fit.
The small form factor would make it easy for a developer to keep one on the (literal) desktop alongside a workstation. Personally, I'd use virtualization instead, but others may prefer having a physical box to play with.
I'm not aware of any systems that do all this for you while still being flexible enough to accommodate lots of unique requirements. The script-based systems I've employed over the years all followed some basic rules:
I've successfully managed hundreds of sites and web apps in this manner, with minimal fuss. Virtualization adds in extra complexity in some respects, but makes other things easier as each VPS can have its own customized environment. As long as everything in dev, test, and prod uses the same base VPS environment, problems should be minimal.
Okay, so how much do you figure it will cost to outfit a server to be capable of supporting an arbitrary number of users, running extremely IO intestive reports on a shared volume whenever they feel like it, including the network infrastructure required to support this? Oh, and don't forget redundancy for the server.
Trust me, I've learned from experience that the local disk space works out to be much cheaper for this sort of thing.
That's a completely difference scenario, and you know it. How about addressing the actual use case in question? He's talking about hundreds of gigabytes, not tens of terabytes. By the way, many organizations would have the "20 TB file server" so everyone can access the data, not necessarily exclusively via the share.
You also didn't specify how often the bulk of your data changes (I'm guessing the bulk of his share doesn't change every day, and yours probably doesn't either). Add in the fact that a 1 TB drive costs about $90 (you can do better if you shop around) and your objection to simply doing incremental updates for a few hundred gigs of data to a local mirror looks pretty ridiculous. It gets even more absurd when you consider how fast you can rsync a crapload of data over even a 100 Mbit connection, let along increasingly prevalent gigabit networks.
Why slam the server with complicated indexing schemes, coupled with multiple users competing for all the data on a potentially frequent basis? That sounds like a much bigger headache than just taking the simple route, unless business requirements specifically stipulate a minute-by-minute ability to run reports on the data in question. Given that the submitter is dealing with very time windows for processing reports in the first place, I don't get the impression that this is the case.
I don't think you understand how rsync works. After the initial mirror, it only transfers changed files, which tends to be a really quick operation. For most organizations, that's going to be a hell of a lot less than the total disk usage on the shared volume.
Whether it defeats the purpose or not depends entirely on the organization's needs. If querying data every few hours in a local app is the objective, that can be met quite effectively with mirroring. Disk space is cheap.
You could just rsync the shared volume to a local drive as frequently as needed and run the search engine on the local copy.
It's quote time:
“All our political forms are exhausted and practically nonexistent. Our parliamentary and electoral system and our political parties are just as futile as dictatorships are intolerable. Nothing is left. And this nothing is increasingly aggressive, totalitarian, and omnipresent. Our experience today is the strange one of empty political institutions in which no one has any confidence any more, of a system of government which functions only in the interests of a political class, and at the same time of the almost infinite growth of power, authority, and social control which makes any one of our democracies a more authoritarian mechanism than the Napoleonic state.”
- Jacques Ellul
Out drinking last night, eh?
The web is not the Internet.
Eternal Sunshine of the Spotless Mind
Price controls are being enforced to prevent what the Finnish government might perceive as "price gouging." This means that no matter how ridiculously remote your location might be, you now have a legal "right" to broadband Internet access at some ISP's hefty expense. That's absurd.
That's not the point at all. The entire point of this story is that the Finnish government has guaranteed access to broadband Internet service to every single citizen in the nation.
I read the article in its entirety; it's ridiculously lacking in details. Unfortunately, so is every other "source" on this topic.
We have two possibilities here: (1) Finland is enforcing price controls (as would seem to be the case to prevent perceived "gouging" by ISPs who now must provide broadband service to anyone, regardless of how remote their location may be), or (2) Finland's government (actually their citizens via taxes, which are very high there) is footing the bill.
Libertarians don't ignore that fact. They simply believe in the principle of "direct harm." There's a huge difference.
Write to them. Call them. Make your views known. If they don't pursue the goals you'd like, perhaps your money would be better donated elsewhere.
That's effectively what Finland's government is doing. They're stipulating that consumers have a legal right to broadband Internet access at a "reasonable" price.
The United States government didn't put price controls on the firearms I own. I purchased them at a value determined by the market.
Okay, I'll rephrase the question. Why should everyone have a legal right to broadband Internet access?
Who pays for this human right of broadband Internet access in Finland? Is it completely subsidized by the government?
You're absolutely right. With Internet access guaranteed, I could warm my bedroom with the waste heat from my computer, cable modem, and router.
... but seriously, how is access to a broadband Internet connection a legal right? Somebody please explain this to me, because the article doesn't give any supporting logic.
I need air to breathe, food to eat, clothes to wear, and a place to sleep at night. As much as I enjoy working in I.T. for a living, I do not need Internet access to survive.
Redirection can be accomplished by any number of means, one of which is simply telling a mail client to BCC an email address for all outgoing mail. File corruption happens all the time, and doesn't necessarily mean the accused had anything to do with it. In fact, outright mass file deletion would be more suspect in my book.
Of course, the odds are extremely good that nobody on Slashdot actually knows the full story, but the "evidence" as presented is absurdly weak for a hacking accusation.
Flyersrights.org, a nonprofit organization founded in 2007, had been investigating surface delays in air travel. According to the suit, Hanni exchanged information with Frederick J. Foreman, who worked for Metron Aviation, which was hired by the U.S. Federal Aviation Administration to study surface delays. The suit says Foreman provided information to Hanni with permission from Metron, including a report that fingered Delta as having excessive surface delays. Metron is also named in the suit.
During the correspondence, AOL informed Hanni that her e-mails, spreadsheets and lists of donors were being redirected to an unknown destination. Also, files on Hanni's computer became corrupted, the suit says. The hacking began in 2008 and continued through this year.
This does not constitute "hacking" (or even cracking, as it should be termed). Unless I've missed something here, the actual allegation is that information was improperly disclosed, but not that an email account was broken into.
Somebody please mod the parent up.