As an aside, reverse engineering was never necessary to understand or duplicate a unix kernel and is therefore his mention of it is a complete red herring.
As a double aside, the Posix specification, which most 1990s-era Unix(tm) clones attempted to meet, was orginally a US Government standard and cannot be copyrighted.
This is probably just another case where some statistician takes a common problem, makes a few guesses, and comes up with some exorbenant figure to scare people into paying attention. You can support any point with statistics.
Not to mention that he probably used a spreadsheet to calculate those statistics...
Well dude, I would have to think the Department of Defense would be able to get the NOAA data even if YOU can't. Remember that the Navy was doing weather forecasts before the Constitution was signed.
Well, let's see. Santorum sold his family's house in Pennsylvania about 7 years ago, bought a shell house in a lower-middle-class community, and last year sent that community a bill for $100,000 for the tuition for his children in Virgina schools. Now, this is apparently legal under PA law, but I guess if he were my Senator it would bother me a bit.
I do agree that the clueless one here is not Santorum, but in this case I don't agree it is Slashdot readers.
I wish people would stop recommending that "people write their congress rep" eveything an innane law pops up. These people don't care, hell I bet most don't even read their own mail. With these web based forms and email, I'm sure peoples' opinions are a dime a dozen and most are immediatly filed promptly into/dev/null.
Um, no. Some congressmen have staffers who actually read letters; others don't. But they all at least count the number of letters they get on a particular topic. And I have talked to staffers who have told me that from time to time a letter actually does make a difference.
Now, using the web forms and e-mail is probably useless. You need to print it out, sign it in blue ink, put a stamp on it, and mail it. Which very few slashdotters will ever do.
In this particular case, it would be more-or-less my decision due to the size of the organization and my level of influence. That would not have been the case at my last job.
But I bring it up because it is a good point: small and mid-sized businesses often have very small IT staffs, much smaller than you would think. In the long-ago days when I worked for Megacorp we would have put together a team to study and research that "Linux thing" that would have been larger than my whole staff today; my current company's IT-to-gross-income ratio is laughable by Mega's standards. Or dozens of times more efficient, whichever you prefer;-)
Again, I think you have an unrealistic view of both the competitive environment and financial structure of non-technology firms. Many of the mfg companies in our local ERP user group are running the app under Windows 95 on PII 266s. The app works just fine and they haven't spent any money on the infrastructure since 1998. That is how a mfg company makes money.
And as I said elsewhere in the thread, there seems to be an unstated assumption that "Linux = better". The fact that Wintel is the most popular platform also means that there are the most techs available for it, that it is well understood, more apps (compare app list for Wintel vs Mac), fewest compatibiity problems with business partners, etc. You are assuming huge superiority for Linux and arguing from there. I would like to move to Linux myself but I don't agree that your premise is a given.
No, the price is reasonable. Especially because you don't need anywhere as near as many servers as you do with Exchange. But our testing is not complete so don't take my word for it - they have a free test version.
I think there is much truth in your post, but you left out one: the mid-sized business has done a careful analysis and determined there is no business benefit to changing its business software or its platform. The basis of this discussion seems to be the unstated premise that using Linux is of course better and cheaper for every business, but perhaps that is not always so?
I don't entirely disagree with what you are saying, but I think it is a bit of an unrealistic view of how non-technology companies work. Having a well-understood business management application that works for you is not considered being "locked in" necessarily, although there is that risk. Having that application require the most most popular business computing platform in the world isn't even on management's radar screen as a competitive issue.
Pardon my asking, but what Windows API was in existence in the late 80s/early 90s? Are you really still using Windows 3.1? (Or was that even around in late 80s?) Alas, by my ignorance of history, I betray my youth.
Most of them were written to what we now call the Win16 APIs: Windows 3.1, 3.0 and in a few cases Windows 386 and its predecessors. Or ported from DOS to Windows 3.1 using some sort of CLI-to-GUI tool. They are slowly moving toward the Win32 API (and I yell at my vendors every week to recompile for Linux), but that isn't an easy transition for the app vendor either for what amounts to an "embedded business" app.
> Not to go crazy on this one, but what is the big > deal is requiring your calendar and address book > be tied to your email client. I guess somewhere > along the line everyone got mixed up and decided > this is the way life should be.
That is your opinion, which is fine. I personally prefer a Unix(tm)-type command-line-based e-mail app myself. But please recognize that 66% or more of Western office workers do not share your opinion, which is a large market.
> firewall is not something that has to be > administered everyday (when it is working > properly).
Yes, but when it does have to be administered the person doing it has the know the system it is running on inside and out, to determine the consequences/side-effects of the changes. That depth of knowledge is not acquired casually.
> But I find it hard to believe that they are > not considering Linux.
We are. I would love to convert our shop to all-Linux tomorrow.
Problem is, mid-sized business use midrange apps. E.g. Forth Shift, Visual Manufacturing, etc. Which were written in the late 80s/early 90s. For the Windows API.
Changing out a business management system is no trivial task for a midsized company, and not undertaken lightly. If and when these midrange vendors compile Linux versions (web-based really doesn't work for high-volume ERP transactions), then we can look at moving to Linux. But unfortunately not before.
Sorry that my writing was unclear: one year, three different treatments! No sooner would I get my capital budget stabilized then BAM!: U-turn.
I concluded that things weren't going well, which they weren't, but given the nature of the business it wasn't going to collapse and nothing illegal was going on. Stupid perhaps!
you need a carefully-planned methodology, not a bunch of yammering 'experts' giving you off-the-cuff advice on a very complex problem...
I once worked for a Fortune 100 company that had a very good accounting staff and formal training in financial theory for engineers who were coming up through the ranks - and one year we peaked at three flip-flops between "buy PCs" and "lease PCs". Each one accompanied by a very authoritative letter from a Tax God proclaiming that This was the correct treatement.
Reality was the company's cash position was fluctuating so badly they had to keep going out and getting new experts to contradict the last ones so they could do what they needed to do.
So don't be so sure the financial wizards really know what they are talking about either!
Windows NT was a better product than Netware in just about any way imaginable. I remember when I made the switch in my career from Netware to NT. I can't think of anything that Netware did better than Windows NT. Netware pretty much sucked ass...
I must admit that it was very early in my IT career that I made this switch. Perhaps my inexperience in Netware had something to do with my opinion of it.
Sorry to have to inform you but you are dead wrong here - except for your last sentence!
Netware 3.11 was a technical marvel for its day, is still in use worldwide, and still has plenty of features that have never been duplicated in NT. 4.11 and its descendents are so far superior to NT in network funcationality that there really isn't any comparision.
Admittedly, Novell blew it on the application server side, and that was one of the things that really hurt vs. NT. I remember the Microsoft salesmen demo'ing NT to business unit managers (over the heads of IS) and touting how it could perform all the network functions and handle app serving. Of course, it couldn't, nor did it really have the network functionality, nor was it cheaper, but it looked as if if it did/was, and that was all that mattered.
Also true that Novell has a long learning curve, and those who gave up early never understood its power. Sort of like Linux;-). But now that I am in an NT environment, not a day goes by that I don't miss some feature of Netware.
Fire up AutoCAD. 10 minutes, done. No more getting out the eraser and spending a month fixing a few lines.
I worked in engineering offices in the drafting-pencil-and-typewriter era. Then I worked on implementing AutoCAD and word processing (for spec writing) at several offices. And finally worked in engineering again after the transition.
There is no question that task efficiency has been improved. The open question in my mind is whether task effectiveness or overall organizational efficiency has improved. When making an engineering change required a 3 day train ride to Chicago and several hours of work at the drafting board, (a) people worked harded to ensure changes didn't need to be made (b) when a change was required, effort was expended to ensure it was done right and the minimum number of times. Same with specs - when every change required you go to beg the evil dept secretary to type a new page, you got the page right in as few tries as possible.
Today people create 37 versions and blast out 283 copies for comment via e-mail. The task productivity is high and the appearance of "action" and "improvement" is great - but is the organization any better off?
I am watching teenagers go through this transition from schoolwork to research right now, and to me the same questions apply. Yeah, Google gets quick results, and I use it a lot myself. But are the results as good as spending some hours in the library actually reading books and papers on the subject? I don't think so.
First you have to convince me that PCs have improved the productivity of the American business/office. The evidence on that one is at best inconclusive. (in your analysis don't forget to factor in the hordes of PC support techs who did not exist as late as 1992 in most businesses).
If you can point to one roadmap that says one consistent thing I would be obliged! The roadmaps I have seen over the last 3 years have said 29 contradictory things, often within the same document.
That question is answered in the very latest post in her blog.
I am sure you realize this, but one of the purposes of a Slashdot interview is to start (or terminate) a "dialogue" with a community when that community feels alienated or ill-informed. Many of us have read that blog and similar statements from MozFo insiders, but would like a bit more in the way of explanation.
I think your application of those reasons definitely needs to be categorized in the "How to Alienate Your Base Users in 30 minutes or less" department. It's like you filed for a divorce, split up the property, moved everything out, and finally after the dust settled, you informed your wife.
Slashdot Mirror
sPh
sPh
sPh
I do agree that the clueless one here is not Santorum, but in this case I don't agree it is Slashdot readers.
sPh
sPh
Now, using the web forms and e-mail is probably useless. You need to print it out, sign it in blue ink, put a stamp on it, and mail it. Which very few slashdotters will ever do.
sPh
In this particular case, it would be more-or-less my decision due to the size of the organization and my level of influence. That would not have been the case at my last job.
;-)
But I bring it up because it is a good point: small and mid-sized businesses often have very small IT staffs, much smaller than you would think. In the long-ago days when I worked for Megacorp we would have put together a team to study and research that "Linux thing" that would have been larger than my whole staff today; my current company's IT-to-gross-income ratio is laughable by Mega's standards. Or dozens of times more efficient, whichever you prefer
sPh
Again, I think you have an unrealistic view of both the competitive environment and financial structure of non-technology firms. Many of the mfg companies in our local ERP user group are running the app under Windows 95 on PII 266s. The app works just fine and they haven't spent any money on the infrastructure since 1998. That is how a mfg company makes money.
And as I said elsewhere in the thread, there seems to be an unstated assumption that "Linux = better". The fact that Wintel is the most popular platform also means that there are the most techs available for it, that it is well understood, more apps (compare app list for Wintel vs Mac), fewest compatibiity problems with business partners, etc. You are assuming huge superiority for Linux and arguing from there. I would like to move to Linux myself but I don't agree that your premise is a given.
sPh
No, the price is reasonable. Especially because you don't need anywhere as near as many servers as you do with Exchange. But our testing is not complete so don't take my word for it - they have a free test version.
sPh
I think there is much truth in your post, but you left out one: the mid-sized business has done a careful analysis and determined there is no business benefit to changing its business software or its platform. The basis of this discussion seems to be the unstated premise that using Linux is of course better and cheaper for every business, but perhaps that is not always so?
sPh
Take a look at Scalix - in our testing so far it seems to integrate with Outlook.
sPh
sPh
sPh
> Not to go crazy on this one, but what is the big
> deal is requiring your calendar and address book
> be tied to your email client. I guess somewhere
> along the line everyone got mixed up and decided
> this is the way life should be.
That is your opinion, which is fine. I personally prefer a Unix(tm)-type command-line-based e-mail app myself. But please recognize that 66% or more of Western office workers do not share your opinion, which is a large market.
sPh
> firewall is not something that has to be
> administered everyday (when it is working
> properly).
Yes, but when it does have to be administered the person doing it has the know the system it is running on inside and out, to determine the consequences/side-effects of the changes. That depth of knowledge is not acquired casually.
sPh
> But I find it hard to believe that they are
> not considering Linux.
We are. I would love to convert our shop to all-Linux tomorrow.
Problem is, mid-sized business use midrange apps. E.g. Forth Shift, Visual Manufacturing, etc. Which were written in the late 80s/early 90s. For the Windows API.
Changing out a business management system is no trivial task for a midsized company, and not undertaken lightly. If and when these midrange vendors compile Linux versions (web-based really doesn't work for high-volume ERP transactions), then we can look at moving to Linux. But unfortunately not before.
sPh
I concluded that things weren't going well, which they weren't, but given the nature of the business it wasn't going to collapse and nothing illegal was going on. Stupid perhaps!
sPh
Reality was the company's cash position was fluctuating so badly they had to keep going out and getting new experts to contradict the last ones so they could do what they needed to do.
So don't be so sure the financial wizards really know what they are talking about either!
sPh
> Leasing on the other hand falls under
> "liabilities".
You can be forced to treat leases as capitalized items if they are found to fall under the concept of "capitalized lease".
sPh
Netware 3.11 was a technical marvel for its day, is still in use worldwide, and still has plenty of features that have never been duplicated in NT. 4.11 and its descendents are so far superior to NT in network funcationality that there really isn't any comparision.
Admittedly, Novell blew it on the application server side, and that was one of the things that really hurt vs. NT. I remember the Microsoft salesmen demo'ing NT to business unit managers (over the heads of IS) and touting how it could perform all the network functions and handle app serving. Of course, it couldn't, nor did it really have the network functionality, nor was it cheaper, but it looked as if if it did/was, and that was all that mattered.
Also true that Novell has a long learning curve, and those who gave up early never understood its power. Sort of like Linux ;-). But now that I am in an NT environment, not a day goes by that I don't miss some feature of Netware.
sPh
There is no question that task efficiency has been improved. The open question in my mind is whether task effectiveness or overall organizational efficiency has improved. When making an engineering change required a 3 day train ride to Chicago and several hours of work at the drafting board, (a) people worked harded to ensure changes didn't need to be made (b) when a change was required, effort was expended to ensure it was done right and the minimum number of times. Same with specs - when every change required you go to beg the evil dept secretary to type a new page, you got the page right in as few tries as possible.
Today people create 37 versions and blast out 283 copies for comment via e-mail. The task productivity is high and the appearance of "action" and "improvement" is great - but is the organization any better off?
I am watching teenagers go through this transition from schoolwork to research right now, and to me the same questions apply. Yeah, Google gets quick results, and I use it a lot myself. But are the results as good as spending some hours in the library actually reading books and papers on the subject? I don't think so.
sPh
First you have to convince me that PCs have improved the productivity of the American business/office. The evidence on that one is at best inconclusive. (in your analysis don't forget to factor in the hordes of PC support techs who did not exist as late as 1992 in most businesses).
sPh
> and the roadmap has said so all the time.
If you can point to one roadmap that says one consistent thing I would be obliged! The roadmaps I have seen over the last 3 years have said 29 contradictory things, often within the same document.
sPh
> IMO the roadmap doesn't make it clear enough that
;-)
> the 1.8 and 1.9 releases shown are platform
> releases and not suite releases.
The four milestone releases* of Suite 1.8 were a bit confusing too
sPh
* I don't think "alpha" and "beta" relate well to the MozFo development process anymore.
Or as WR put it:
sPh