Documentation of your systems is just as important as your code when it comes to maintaining systems you've built. So yeah, you can either: 1) convince your boss to hire a technical writer; 2) write the documentation yourself; 3) Complain to slashdot about how you're so critical to your ex-employers that they won't leave you alone, even after you've left the company.
If you want to give free work to people, be my guest. If you enjoy being continually hassled with questions by your replacement, don't write anything down. If you want to do a thorough, professional job, you write some documentation of your systems - for yourself, and for your eventual replacement, and you turn that documentation over when you leave the company, with the "$200/hr consulting fee" message.
Often times in my experience, the engineers who refuse to write documentation also write the worst & most brittle code, because they're making it up as they go along, rather than thinking about the problem they're trying to solve, and how they want to solve it, before they start churning out objects and methods.
It makes me so important to the companies that they start having problems letting go when I decide to leave
Where's your documentation? Your replacement should be able to refer to that and figure out what you did.
If an ex-employer keeps calling after you've left, explain to them patiently that "after date X, if you want help, it'll cost you $200 per hour for consultation, my minimum billing increment is 1 hour, and I will not agree to provide more than N hours of support per week to you on my off-hours."
If you try being a nice guy, they will take advantage of it, and not bother to figure out things on their own. If they know that calling you to answer a single question is going to cost them $200 for the 5 minutes it takes you to have the conversation on the phone, they'll have more incentive to figure out the answer before wasting your time.
The potential for conflict in (or involving) a major oil producer, sure. That's because they are already producing, and actively pumping that oil into refineries and industry.
The Macondo well was not yet producing, they were wrapping up the exploration phase when the blowout occurred, so the only effect it would have had on the commodity markets would be a rise in prices of oil futures, not current production prices. If the well had blown up while in production, this might have moved the prices more.
Today's crude oil price is ~$75 - 80 per barrel. 4.9 million barrels * $80 per barrel = ~ $368 million worth of oil spilled, at today's prices. That's a LOT, but it's a drop in the bucket compared to the revenues the oil producers log.
If they were spending all that money exploring & drilling at 5k+ meters of depth for a mere 4.9 million barrels, then we're much closer to peak oil & an economic collapse on account of energy crisis than anybody has suggested.
Aesthetically, a choice to de-emphasize the "ZOMG LOOKIT I GOTS SCROLLBARS AND TITLE BAR AND ICONZZZZ" in favor of letting the content in your apps themselves take center stage is not *necessarily* a bad thing.
If the "plumbing" of the OS & the window manager is overshadowing the content in the apps to the point of distracting users, I'd say that toning down the window manager is probably a reasonable decision from a design standpoint. I don't know if there's anything to indicate how the move towards more and more gray-scale in OS X affects usability, but the consistent & muted layout for the window manager helps minimize distractions for me, so I don't mind it so much. Given Apple's focus on usability, I wouldn't be surprised if they had *some* data indicating that this was the case, but I'm not aware of any published studies on this.
That said, I'm not primarily a mouse-clicker, I like my keyboard tools like Quicksilver for handling many routine tasks and commonly-used apps, so if you asked me which icon was which from my apps folders, I would probably be hard pressed to identify some of them.
Vendor Lock-in is far from weasel (your words) and more about expensive, costly, prohibitive, limiting, restricted, etc...
Yes, I called it a weasel word, because it *IS* a weasel word. Please explain to me how buying a Microsoft product is more "prohibitive, limiting, restricted, etc." without using the "Well if Microsoft goes out of business tomorrow, and suddenly every copy of Windows and Office and SQL server stops working, you'll be totally screwed" fairies-and-moondust argument.
What part of M$ TCO being infinite, unending, forever don't you understand.
Oh, I see - you must have missed the part where I explained that the argument is a fucking ridiculous pedantic claim because: 1) Linux requires ongoing support and maintenance of hardware & systems, just like Windows. 2) MSFT licensing is a very small component of the TCO of any computer system;
So by that, if you want to claim Microsoft TCO is "infinite," well then, so is Linux's. TCO is not just "what you spend to get a copy of the software," and you're an idiot if you don't understand that, or a liar if you do.
Every business I worked for, managed and owned expected some return on their investment, even in IT...so owning something gives you a return, perhaps a small one, but still a return on your investment.
If you are buying computer systems, and expecting the ROI to come from "we own the software on these things," your company is doomed to failure. The ROI from IT comes from increased efficiency, scalability, and automation of your business processes. Not from capital expenditures on hardware that is depreciated over 3 years. There is no expectation of "return" on the purchase of the hardware, there is an expectation that the expense of purchasing the systems will be offset in decreased spending in other areas, or increased revenues as a result of purchasing the servers. NO company buys a new server and goes, "Man, we're gonna wait 3 years, and once the value of these things goes up, BAM! We'll have cornered the market."
You won't own anything; you can't even sell the PC with the software. There's no ownership.
Then you are making a pointless semantic argument. The business doesn't care whether or not it "owns" the operating system, it cares whether or not the operating system allows the employees of the business to get their jobs done and make a profit, and is mostly concerned with *maximizing* that profit. If it can get more profit out of cheap windows admins and cheap OS licenses, it will do so. If it makes you feel better, read "TCO" as the "total cost of operation" rather than "ownership".
Converting an existing enterprise to Linux costs a significant amount of money, time, and manpower. That all has a dollar cost. If there is no compelling reason other than "But it's OPEN!" to shift, why would they spend the money to do so? Where will they recoup that investment? Try to answer without weasel words like "vendor lock-in" and "freedom."
Not exactly little, but then also maybe not that much -- that's difficult to assess from a couple of words, don't you think?;-]
I think it's pretty evident that "not that much" is likely the case from the comments you've made here. Large enterprises simply do not run their IT departments the way you suggest, so the only reasonable conclusion to draw is that you have litte-to-no experience in the enterprise space.
Linux certainly has a place in the enterprise, but it's foolish to suggest that it's automatically cheaper to put it everywhere because "a copy of Linux is free."
Yeah, do you really want to be the one running around proclaiming the year of DILDOS on the desktop, or DILDOS on the phone, or DILDOS on the camera? Or heaven forbid, motherfucking DILDOS on a motherfucking plane?
I just don't think I'd really care to work in an IT shop that used nothing but DILDOS... IT is a male dominate field, we have more than enough dicks without importing fake ones.
Because my boss wanted to get the intern set up quickly, so I suggested what I knew to be a "quick / decent" tool for them to do some image editing in.
Whether or not it got used is irrelevant to me, but it is relevant to this discussion, where somebody was claiming that nobody would refuse to use it based on a poor name choice. I can offer at least one example of a time that it *did* get passed over for exactly that reason.
"appeal to internet dictionary"? Dude, it's called a citation - it's not just a slang reference that somebody picked up from Pulp Fiction, 'gimp' is a legitimate word, referring to a limp. That's all that was intended to demonstrate - I'm not saying those other definitions aren't also legitimate, but let's be honest: ask somebody to define a "gimp", and see what the overwhelming response is - hint: most people aren't horsemen, or tailors.
As far as naming a company, why should I be bothered? I never claimed that 'many companies are refusing to use it,' I simply provided a single anecdote where at least my boss opted not to, for apparently no other reason than that he thought the name sounded unprofessional. You can refuse to believe it if you want, but you can't say the phenomenon doesn't exist: terrible branding gives a bad first impression.
Yeah, I'll get right on wasting my time worrying about the branding of some software that I don't care a thing about. If you don't want people calling it "GIMP", don't name it "GIMP."
If you're saying that "GNU Image Processor" would be a "friendlier" name, then you're also implying that the term 'GIMP' will, at the very least raise a few eyebrows or cause a few snickers. And if that's the reaction you want for your software, that's fine. But if you want it to be seen as a viable alternative to a well-known proprietary package, then you don't hobble it with a terrible "hurr hurr you said dick" type of name.
Funny, my company uses Apache, Tomcat, Perl, Python, PHP, Linux, GCC, Java, Eclipse, and a host of other FOSS software & tools. This has nothing to do with licensing or purchasing policies.
This was me, suggesting something to my manager, after which he looked at me like I had suggested we engage in a little light tickling - just for fun! - and then said, "No, let's just use Photoshop." The name gets an immediate response, and it's sometimes not a positive one.
If you choose a bad name for your software, this is going to happen. You won't be banned outright, managers will simply say, "No, let's just wait and get something professional." If Linux had named itself "DILDOS" (Dynamic Interactive Logical Display Operating System!), it probably wouldn't have gained much traction either, simply due to the snicker factor.
Definition 3 also happens to be the one most people are familiar with. GIMP is a *horrible* name for that reason, and I can tell you that, while not banned, I had a manager look at me and say, "No, we'll just buy Photoshop," one time when I suggested "Why not just grab a copy of Gimp from the web to get the intern working on some of these images you want?"
Of course it does, and I didn't say that those two items are the only components of TCO.
I was illustrating a point, where somebody basically claimed that, because of "license fees," the "TCO of Windows" was infinite, and that Linux was clearly the lower TCO. Anybody with a shred of common sense knows that license fees are a vanishingly small portion of TCO, and it's easy to see that simply the cost of labor to deploy, maintain, and manage could offset the license costs.
As far as: "How about the costs incurred by vendor lock-in?", converseley, how about the costs incurred by having to develop your own solution because no pre-existing software for Linux is available?
Let's not pretend that Windows has only negatives and Linux has only positives.
And how much does a *comparable*, unlocked, 32GB Android phone cost from Motorola, HTC, or Samsung? Not the cheapest phone that runs Android, mind you - one comparable to an iPhone 4.
Why would it be worth mentioning? The story is about the first Google Voice app to hit the App Store, not about "the first time a Google Voice app is available!"
The Cydia version of the GV app wasn't the first time GV was available, either, was it? So why does Cydia deserve special mention?
Linux requires ongoing support and management, just like windows.
If cost of windows licenses plus cost of support staff for ongoing operations is less than cost of Linux (you can buy support from companies like red hat, and guess what an enterprise is going to do?) plus the cost of support staff for ongoing operations, then tco of windows is LOWER than tco of Linux.
Frankly, with enterprise licensing agreements, I'd be surprised if the licenses were anything approaching even a sizable minority, much less a majority, of an organization's total cost of ownership for windows. I suspect you have precious little experience in large enterprise if you're making these claims.
Fwiw, my company uses centrifydc, and has been pretty happy with the results. Largish (~35k users) enterprise, aix, solaris, Linux, and lots of winxp & server. As someone who primarily works on the Linux/unix side of things, I was skeptical about the whole "AD integration!" aspect, but it's been a pretty solid tool. Only noticeable hassle is I no longer have a sudo to unlock a user's account if they fat finger their password.:). We did have to upgrade a couple of our samba servers to a centrify-compatible one, but that was pretty straightforward and the rollout team provided us with the centrify-specific bits of smb.conf.
Slashdot Mirror
Documentation of your systems is just as important as your code when it comes to maintaining systems you've built. So yeah, you can either:
1) convince your boss to hire a technical writer;
2) write the documentation yourself;
3) Complain to slashdot about how you're so critical to your ex-employers that they won't leave you alone, even after you've left the company.
If you want to give free work to people, be my guest. If you enjoy being continually hassled with questions by your replacement, don't write anything down. If you want to do a thorough, professional job, you write some documentation of your systems - for yourself, and for your eventual replacement, and you turn that documentation over when you leave the company, with the "$200/hr consulting fee" message.
Often times in my experience, the engineers who refuse to write documentation also write the worst & most brittle code, because they're making it up as they go along, rather than thinking about the problem they're trying to solve, and how they want to solve it, before they start churning out objects and methods.
I'm sorry... you were saying something about being shoehorned into a stereotype by people who don't understand you? I lost you in all the irony.
Where's your documentation? Your replacement should be able to refer to that and figure out what you did.
If an ex-employer keeps calling after you've left, explain to them patiently that "after date X, if you want help, it'll cost you $200 per hour for consultation, my minimum billing increment is 1 hour, and I will not agree to provide more than N hours of support per week to you on my off-hours."
If you try being a nice guy, they will take advantage of it, and not bother to figure out things on their own. If they know that calling you to answer a single question is going to cost them $200 for the 5 minutes it takes you to have the conversation on the phone, they'll have more incentive to figure out the answer before wasting your time.
The potential for conflict in (or involving) a major oil producer, sure. That's because they are already producing, and actively pumping that oil into refineries and industry.
The Macondo well was not yet producing, they were wrapping up the exploration phase when the blowout occurred, so the only effect it would have had on the commodity markets would be a rise in prices of oil futures, not current production prices. If the well had blown up while in production, this might have moved the prices more.
Today's crude oil price is ~$75 - 80 per barrel. 4.9 million barrels * $80 per barrel = ~ $368 million worth of oil spilled, at today's prices. That's a LOT, but it's a drop in the bucket compared to the revenues the oil producers log.
If they were spending all that money exploring & drilling at 5k+ meters of depth for a mere 4.9 million barrels, then we're much closer to peak oil & an economic collapse on account of energy crisis than anybody has suggested.
How about you go read the LA Times article linked in the summary, after which you can answer that question for yourself?
When did Ubuntu announce a new release?
Aesthetically, a choice to de-emphasize the "ZOMG LOOKIT I GOTS SCROLLBARS AND TITLE BAR AND ICONZZZZ" in favor of letting the content in your apps themselves take center stage is not *necessarily* a bad thing.
If the "plumbing" of the OS & the window manager is overshadowing the content in the apps to the point of distracting users, I'd say that toning down the window manager is probably a reasonable decision from a design standpoint. I don't know if there's anything to indicate how the move towards more and more gray-scale in OS X affects usability, but the consistent & muted layout for the window manager helps minimize distractions for me, so I don't mind it so much. Given Apple's focus on usability, I wouldn't be surprised if they had *some* data indicating that this was the case, but I'm not aware of any published studies on this.
That said, I'm not primarily a mouse-clicker, I like my keyboard tools like Quicksilver for handling many routine tasks and commonly-used apps, so if you asked me which icon was which from my apps folders, I would probably be hard pressed to identify some of them.
Yes, I called it a weasel word, because it *IS* a weasel word. Please explain to me how buying a Microsoft product is more "prohibitive, limiting, restricted, etc." without using the "Well if Microsoft goes out of business tomorrow, and suddenly every copy of Windows and Office and SQL server stops working, you'll be totally screwed" fairies-and-moondust argument.
Oh, I see - you must have missed the part where I explained that the argument is a fucking ridiculous pedantic claim because:
1) Linux requires ongoing support and maintenance of hardware & systems, just like Windows.
2) MSFT licensing is a very small component of the TCO of any computer system;
So by that, if you want to claim Microsoft TCO is "infinite," well then, so is Linux's. TCO is not just "what you spend to get a copy of the software," and you're an idiot if you don't understand that, or a liar if you do.
If you are buying computer systems, and expecting the ROI to come from "we own the software on these things," your company is doomed to failure. The ROI from IT comes from increased efficiency, scalability, and automation of your business processes. Not from capital expenditures on hardware that is depreciated over 3 years. There is no expectation of "return" on the purchase of the hardware, there is an expectation that the expense of purchasing the systems will be offset in decreased spending in other areas, or increased revenues as a result of purchasing the servers. NO company buys a new server and goes, "Man, we're gonna wait 3 years, and once the value of these things goes up, BAM! We'll have cornered the market."
Then you are making a pointless semantic argument. The business doesn't care whether or not it "owns" the operating system, it cares whether or not the operating system allows the employees of the business to get their jobs done and make a profit, and is mostly concerned with *maximizing* that profit. If it can get more profit out of cheap windows admins and cheap OS licenses, it will do so. If it makes you feel better, read "TCO" as the "total cost of operation" rather than "ownership".
Converting an existing enterprise to Linux costs a significant amount of money, time, and manpower. That all has a dollar cost. If there is no compelling reason other than "But it's OPEN!" to shift, why would they spend the money to do so? Where will they recoup that investment? Try to answer without weasel words like "vendor lock-in" and "freedom."
I think it's pretty evident that "not that much" is likely the case from the comments you've made here. Large enterprises simply do not run their IT departments the way you suggest, so the only reasonable conclusion to draw is that you have litte-to-no experience in the enterprise space.
Linux certainly has a place in the enterprise, but it's foolish to suggest that it's automatically cheaper to put it everywhere because "a copy of Linux is free."
Yeah, do you really want to be the one running around proclaiming the year of DILDOS on the desktop, or DILDOS on the phone, or DILDOS on the camera? Or heaven forbid, motherfucking DILDOS on a motherfucking plane?
I just don't think I'd really care to work in an IT shop that used nothing but DILDOS... IT is a male dominate field, we have more than enough dicks without importing fake ones.
Because my boss wanted to get the intern set up quickly, so I suggested what I knew to be a "quick / decent" tool for them to do some image editing in.
Whether or not it got used is irrelevant to me, but it is relevant to this discussion, where somebody was claiming that nobody would refuse to use it based on a poor name choice. I can offer at least one example of a time that it *did* get passed over for exactly that reason.
Bah, wrong "Reply" button.
Uh huh. And how much was a comparable android phone selling for at that time, with the same terms & conditions as the iPhone?
And how much was a comparable android phone at that time, with the same terms & conditions as the iPhone?
"appeal to internet dictionary"? Dude, it's called a citation - it's not just a slang reference that somebody picked up from Pulp Fiction, 'gimp' is a legitimate word, referring to a limp. That's all that was intended to demonstrate - I'm not saying those other definitions aren't also legitimate, but let's be honest: ask somebody to define a "gimp", and see what the overwhelming response is - hint: most people aren't horsemen, or tailors.
As far as naming a company, why should I be bothered? I never claimed that 'many companies are refusing to use it,' I simply provided a single anecdote where at least my boss opted not to, for apparently no other reason than that he thought the name sounded unprofessional. You can refuse to believe it if you want, but you can't say the phenomenon doesn't exist: terrible branding gives a bad first impression.
Yeah, I'll get right on wasting my time worrying about the branding of some software that I don't care a thing about. If you don't want people calling it "GIMP", don't name it "GIMP."
If you're saying that "GNU Image Processor" would be a "friendlier" name, then you're also implying that the term 'GIMP' will, at the very least raise a few eyebrows or cause a few snickers. And if that's the reaction you want for your software, that's fine. But if you want it to be seen as a viable alternative to a well-known proprietary package, then you don't hobble it with a terrible "hurr hurr you said dick" type of name.
Funny, my company uses Apache, Tomcat, Perl, Python, PHP, Linux, GCC, Java, Eclipse, and a host of other FOSS software & tools. This has nothing to do with licensing or purchasing policies.
This was me, suggesting something to my manager, after which he looked at me like I had suggested we engage in a little light tickling - just for fun! - and then said, "No, let's just use Photoshop." The name gets an immediate response, and it's sometimes not a positive one.
If you choose a bad name for your software, this is going to happen. You won't be banned outright, managers will simply say, "No, let's just wait and get something professional." If Linux had named itself "DILDOS" (Dynamic Interactive Logical Display Operating System!), it probably wouldn't have gained much traction either, simply due to the snicker factor.
http://dictionary.reference.com/browse/gimp
Definition 3 also happens to be the one most people are familiar with. GIMP is a *horrible* name for that reason, and I can tell you that, while not banned, I had a manager look at me and say, "No, we'll just buy Photoshop," one time when I suggested "Why not just grab a copy of Gimp from the web to get the intern working on some of these images you want?"
It does happen.
Of course it does, and I didn't say that those two items are the only components of TCO.
I was illustrating a point, where somebody basically claimed that, because of "license fees," the "TCO of Windows" was infinite, and that Linux was clearly the lower TCO. Anybody with a shred of common sense knows that license fees are a vanishingly small portion of TCO, and it's easy to see that simply the cost of labor to deploy, maintain, and manage could offset the license costs.
As far as: "How about the costs incurred by vendor lock-in?", converseley, how about the costs incurred by having to develop your own solution because no pre-existing software for Linux is available?
Let's not pretend that Windows has only negatives and Linux has only positives.
And how much does a *comparable*, unlocked, 32GB Android phone cost from Motorola, HTC, or Samsung? Not the cheapest phone that runs Android, mind you - one comparable to an iPhone 4.
Why would it be worth mentioning? The story is about the first Google Voice app to hit the App Store, not about "the first time a Google Voice app is available!"
The Cydia version of the GV app wasn't the first time GV was available, either, was it? So why does Cydia deserve special mention?
Like the man said... "There is no other app store worth mentioning."
Linux requires ongoing support and management, just like windows.
If cost of windows licenses plus cost of support staff for ongoing operations is less than cost of Linux (you can buy support from companies like red hat, and guess what an enterprise is going to do?) plus the cost of support staff for ongoing operations, then tco of windows is LOWER than tco of Linux.
Frankly, with enterprise licensing agreements, I'd be surprised if the licenses were anything approaching even a sizable minority, much less a majority, of an organization's total cost of ownership for windows. I suspect you have precious little experience in large enterprise if you're making these claims.
Fwiw, my company uses centrifydc, and has been pretty happy with the results. Largish (~35k users) enterprise, aix, solaris, Linux, and lots of winxp & server. As someone who primarily works on the Linux/unix side of things, I was skeptical about the whole "AD integration!" aspect, but it's been a pretty solid tool. Only noticeable hassle is I no longer have a sudo to unlock a user's account if they fat finger their password. :). We did have to upgrade a couple of our samba servers to a centrify-compatible one, but that was pretty straightforward and the rollout team provided us with the centrify-specific bits of smb.conf.