I do get the diff between the layers. What I'm trying to suggest is that if there's going to be some rewiring in the fishbowl, why not use the fastest, cheapest solution that has the potential to support all the scenarios? I don't have a 10 Gb/s server room, I have a little 1 Gb/s server room; the SATA & USB specs are either already > than MY twisted pair bandwidth and for a lower cost or will be there in the near future and by trends I've observed will continue to out pace twisted pair bandwidth/$.
Given the direction SATA & USB is going, the rate at which its bandwidth has increased relative to traditional CATx ethernet, and the relatively lower cost of interconnection devices, is Ethernet really the best? If we're going to making significant wiring changes in server rooms I'd prefer to just do it once and standardize on the cheapest, fastest "2-wire" solution that makes sense.
You just know somewhere in a laboratory deep underground in Seattle there's a team of scientists working on sending an advanced cybernetic assassin back in time to locate Linus Torvalds...
Coming Soon: "T4: The Redemption"
Maybe this is just some sort of veiled posturing actually being driven by the fact that using search engines to research law & other court/case activity is threatening business models like Westlaw, Lexis-Nexis. That would be tragic.
Now if we could just get a hiearchical data model and associated standards based query language at the same time (XML, xquery, xupdate, etc) it truly would be Christmas come early. The potential of a FOSS, production ready NXDB is intoxicating (Exist-db, Monet, etc. are sooo close).
If voting's not working (as seems to be the consensus) maybe it's time to escalate matters and take advantage of some of the other provisions in the US Constitution. Now... where did I put my gun?
There's got to be a marketing corollary to the Heisenberg Uncertainty Principal that properly explains the inevitable inaccuracies when modeling consumer behavior.
Hey I'm the first one to scoff at complexity creep. It's usually bad news for business. But some percentage of today's tech toys are tomorrow's sine qua non's... I gotta say I'm feelin' this one.
A lot of people laughed at the concept of a commercial web not too long ago...
Re:nuttin' wrong with growin' your own
on
What NAS To Buy?
·
· Score: 1
$2k for what you describe doesn't conflict w/$4k for what I described. For example:
1. there's a significant diff between an opteron with a big cache and an Athlon 64 (plus I got twice the RAM)
2. the IO controller I used was expensive (8 ch's of 3gb/sec with large on card cache and raid functionality)
3.. I paid a premium for large fast drives that you did not
nuttin' wrong with growin' your own
on
What NAS To Buy?
·
· Score: 1
I've gone through the exact same decision making process you are and made the same decision you did; build your own: 4U rackable box, 750W p/s, dual capable but single CPU Opteron MB w/4 GB RAM, 8 Ch SATA 2 (3 Gb/sec) LSI logic controller, 4x 1TB SATA II drives, 1 Intel Pro dual Gb NIC, OpenSuse 10.3, total $4,000. This box out performed commercially available solutions (closer to network/IO wire speed @ lower cost). This choice always surprises us with its performance and we have lots of options as far as growing the box over its life time.
1. "The Wave of the Future", (C) 1982, Nokes Berry Graphics (poster ordered from a Byte magazine ad in ~1983); a version of a famous Japanese wood carving with a modern CGI twist
2. mass market chart showing the history of computer languages (a freebie from O'Reilly)
3. mass market chart showing the history of operating systems (can't remember source)
4. "The Magnificent Recovery", Ara Hagopian (contemporary artist who's work is reminiscent of fractals or other mathematical images)
5. "Murphy's Law For Computers", poster (circa ~1980)
6. a 24" stone Buddha (purchased from garden supply store)
7. Tibetan prayer flag banner
(the latter two are hung in the server room; while I can't provide concrete data, the assumption is they keep BSOD's at bay on few remaining MS servers)
My Vista notebook, XP x64 (dual core, 2 Gb RAM) laptop took several minutes to boot. No linux laptop I ever used (barring Apple HW- for some reason open firmware introduces a lag when booting linux), even on antiquated HW ever took close to the time that Vista took to boot on modern HW.
Gimp is only a competitor to Photoshop if you're talking about a user about to start learning an image editor.
CMYK issues aside (ie. Gimp native support of CMYK coming but not here yet) If you're faced with the real world task of hiring graphics designers that were trained to use Photoshop in school, then Gimp is not a competitor. The cost of retraining is not trivial- both are complex apps with steep learning curves; especially for non-tech. users.
Don't we already know the answer to your question? If we consider a start time frame of the onset of microcomputer's (ie. c. 1980) FOSS has already lasted as long as it's commercial counterpart and I don't think we have any reason to expect it to start failing.
As other's have pointed out in several ways, losing revenue to competition is not a cost.
Putting that aside for the moment, I do think the soft cost of some of my proprietary sw choices over the years is a true cost. Let's say an IT staff spends 15% of the year dealing with the short comings of proprietary sw... if we tally that across the industry, I bet it's not that small a cost (even relative to imaginary $60 B loss). How much time has been spent dealing with mal-ware due to a particular OS (or its associated email clients) being insecure? I suspect that time represents a HUGE cost.
I used Vista for one month as shipped with a recently purchased laptop but w/a +1 Gb RAM expansion, bringing the 1.8 Ghz Turion 64 X2 laptop to it's max of 2 Gb. Based on this experience my advice is along the lines of the parent post but goes a bit further: Don't install Vista if you need your machine. I won't comment on what's better to use, but if you're looking for a responsive machine with which to get work done, Vista is not the OS for you unless you have access to hardware from the future.
I think it's alright if you follow certain guidelines; for example don't get too technical. Using "seems normal" and "fucking long" instead of elapsed time is probably appropriate for the target market. No need to quibble with actual numbers.
Here's a benchmark I recently conducted: IE6 Seems Normal under Wine but takes too Fucking Long under WinXP (on the same hardware).
I find it interesting that while MS claimed at one point it was too hard to disintegrate IE from Windows, an intrepid group of OSS developers are able to get IE to run in non-MS host OSes. Which do you think is really the more difficult task? Aside: I'm not implying these are equivalent tasks just that one is probably a lot more difficult.
So? We've got a number of years behind us where they've demonstrated they won't protect customer privacy on their own, regardless of what the customer asks for. Why not motivate them with increased revenue? In many markets there's only 1 broadband solution available so, barring legislation, they can pretty much do whatever they want and not risk losing customers. I'd love to see the appropriate legislation in place but my understanding is this happens fairly slowly in the best of circumstances and fighting a battle against what is essentially an entrenched regime with deep pockets (consider the marketing industry and the many ways they've been able to build detailed demographic databases for years) isn't something I'd expect to go smoothly. We're talking about credit card companies and super-retailers like Walmart: resourceful entities with massive resources (ie. lobbying leverage). Our private information has been a commodity to these people for years.
To all you ISP exec's that might be reading this dialog: I'd pay $5/mo more if you'd anonymize my use of the internet (in a way I can verify) and if your service terms stated that I was anonymized in very clear language (ie. no legalese loopholes).
- p
Perhaps 60 Mb is only the portion of the code base that resides in our universe. The rest of the O/S could be running on a cluster of quantum computers beyond the event horizon which communicate via a wormhole with the instance on the user's PC. Hence the name.
Microsoft's future OSes running in an alternative universe is a win-win as far as I'm concerned. In fact, I'd be thrilled if they moved their HQ there. Can we vote them off the planet?
Sorry, sarcasm and ideology aside (and assuming there's no nefarious agenda which there may well be), I think it's great that they've established a group to experiment with new OS concepts. Welcome to the '90's MS.
Anybody think the real agenda here is for Zend to better monetize PHP? I don't know the numbers but my working assumptions are: some big % of web sites use PHP, the vast majority of them are not doing it on Windows, & the vast majority aren't paying Zend anything nor are they going to be receptive to any sort of costs being imposed on them.
On the other hand, if there's a market segment used to paying somebody for their software, and paying every year to keep it running, wouldn't that be a great market segment to expand into if you're looking for revenue? This sounds to me like a solid motivation for Zend to better support the Windows world (or at least issue PR statements to this effect). Side note: I'm not suggesting that Zend will get more money out of PHP users on Win directly, but MS will probably get more server rev. if PHP is better supported on Win, which might mean that MS provides some incentive to Zend.
I think every once in awhile some executive on the fringe of the FOSS world gets a hair up his ass and assumes he can monetize some market leading piece of FOSS- completing missing the point that the market share was probably enabled by two things: quality code thats does something useful and available at no/low cost. As soon as you try to siphon off revenue, you're going to upset the applecart and the result may be a decline in market share. Sorry- free rides are pretty hard to come by no matter how clever you are with smoke and mirrors or bait & switch.
Slashdot Mirror
I certainly have a great deal of respect for Mr. Mcnealy, but I'm not sure that includes expecting him to objectively comment on MS's competition.
I anxiously await his analysis. :-)
I do get the diff between the layers. What I'm trying to suggest is that if there's going to be some rewiring in the fishbowl, why not use the fastest, cheapest solution that has the potential to support all the scenarios? I don't have a 10 Gb/s server room, I have a little 1 Gb/s server room; the SATA & USB specs are either already > than MY twisted pair bandwidth and for a lower cost or will be there in the near future and by trends I've observed will continue to out pace twisted pair bandwidth/$.
Given the direction SATA & USB is going, the rate at which its bandwidth has increased relative to traditional CATx ethernet, and the relatively lower cost of interconnection devices, is Ethernet really the best? If we're going to making significant wiring changes in server rooms I'd prefer to just do it once and standardize on the cheapest, fastest "2-wire" solution that makes sense.
You just know somewhere in a laboratory deep underground in Seattle there's a team of scientists working on sending an advanced cybernetic assassin back in time to locate Linus Torvalds ...
Coming Soon: "T4: The Redemption"
Maybe this is just some sort of veiled posturing actually being driven by the fact that using search engines to research law & other court/case activity is threatening business models like Westlaw, Lexis-Nexis. That would be tragic.
Now if we could just get a hiearchical data model and associated standards based query language at the same time (XML, xquery, xupdate, etc) it truly would be Christmas come early. The potential of a FOSS, production ready NXDB is intoxicating (Exist-db, Monet, etc. are sooo close).
Maybe we should be watching to see if he's blinking anything in Morse Code...
If voting's not working (as seems to be the consensus) maybe it's time to escalate matters and take advantage of some of the other provisions in the US Constitution. Now ... where did I put my gun?
There's got to be a marketing corollary to the Heisenberg Uncertainty Principal that properly explains the inevitable inaccuracies when modeling consumer behavior.
Can't we, as citizens, vote?
A lot of people laughed at the concept of a commercial web not too long ago...
$2k for what you describe doesn't conflict w/$4k for what I described. For example: 1. there's a significant diff between an opteron with a big cache and an Athlon 64 (plus I got twice the RAM) 2. the IO controller I used was expensive (8 ch's of 3gb/sec with large on card cache and raid functionality) 3.. I paid a premium for large fast drives that you did not
I've gone through the exact same decision making process you are and made the same decision you did; build your own: 4U rackable box, 750W p/s, dual capable but single CPU Opteron MB w/4 GB RAM, 8 Ch SATA 2 (3 Gb/sec) LSI logic controller, 4x 1TB SATA II drives, 1 Intel Pro dual Gb NIC, OpenSuse 10.3, total $4,000. This box out performed commercially available solutions (closer to network/IO wire speed @ lower cost). This choice always surprises us with its performance and we have lots of options as far as growing the box over its life time.
1. "The Wave of the Future", (C) 1982, Nokes Berry Graphics (poster ordered from a Byte magazine ad in ~1983); a version of a famous Japanese wood carving with a modern CGI twist 2. mass market chart showing the history of computer languages (a freebie from O'Reilly) 3. mass market chart showing the history of operating systems (can't remember source) 4. "The Magnificent Recovery", Ara Hagopian (contemporary artist who's work is reminiscent of fractals or other mathematical images) 5. "Murphy's Law For Computers", poster (circa ~1980) 6. a 24" stone Buddha (purchased from garden supply store) 7. Tibetan prayer flag banner (the latter two are hung in the server room; while I can't provide concrete data, the assumption is they keep BSOD's at bay on few remaining MS servers)
My Vista notebook, XP x64 (dual core, 2 Gb RAM) laptop took several minutes to boot. No linux laptop I ever used (barring Apple HW- for some reason open firmware introduces a lag when booting linux), even on antiquated HW ever took close to the time that Vista took to boot on modern HW.
Gimp is only a competitor to Photoshop if you're talking about a user about to start learning an image editor. CMYK issues aside (ie. Gimp native support of CMYK coming but not here yet) If you're faced with the real world task of hiring graphics designers that were trained to use Photoshop in school, then Gimp is not a competitor. The cost of retraining is not trivial- both are complex apps with steep learning curves; especially for non-tech. users.
Don't we already know the answer to your question? If we consider a start time frame of the onset of microcomputer's (ie. c. 1980) FOSS has already lasted as long as it's commercial counterpart and I don't think we have any reason to expect it to start failing.
As other's have pointed out in several ways, losing revenue to competition is not a cost. Putting that aside for the moment, I do think the soft cost of some of my proprietary sw choices over the years is a true cost. Let's say an IT staff spends 15% of the year dealing with the short comings of proprietary sw... if we tally that across the industry, I bet it's not that small a cost (even relative to imaginary $60 B loss). How much time has been spent dealing with mal-ware due to a particular OS (or its associated email clients) being insecure? I suspect that time represents a HUGE cost.
I used Vista for one month as shipped with a recently purchased laptop but w/a +1 Gb RAM expansion, bringing the 1.8 Ghz Turion 64 X2 laptop to it's max of 2 Gb. Based on this experience my advice is along the lines of the parent post but goes a bit further: Don't install Vista if you need your machine. I won't comment on what's better to use, but if you're looking for a responsive machine with which to get work done, Vista is not the OS for you unless you have access to hardware from the future.
Here's a benchmark I recently conducted: IE6 Seems Normal under Wine but takes too Fucking Long under WinXP (on the same hardware).
I find it interesting that while MS claimed at one point it was too hard to disintegrate IE from Windows, an intrepid group of OSS developers are able to get IE to run in non-MS host OSes. Which do you think is really the more difficult task? Aside: I'm not implying these are equivalent tasks just that one is probably a lot more difficult.
So? We've got a number of years behind us where they've demonstrated they won't protect customer privacy on their own, regardless of what the customer asks for. Why not motivate them with increased revenue? In many markets there's only 1 broadband solution available so, barring legislation, they can pretty much do whatever they want and not risk losing customers. I'd love to see the appropriate legislation in place but my understanding is this happens fairly slowly in the best of circumstances and fighting a battle against what is essentially an entrenched regime with deep pockets (consider the marketing industry and the many ways they've been able to build detailed demographic databases for years) isn't something I'd expect to go smoothly. We're talking about credit card companies and super-retailers like Walmart: resourceful entities with massive resources (ie. lobbying leverage). Our private information has been a commodity to these people for years.
To all you ISP exec's that might be reading this dialog: I'd pay $5/mo more if you'd anonymize my use of the internet (in a way I can verify) and if your service terms stated that I was anonymized in very clear language (ie. no legalese loopholes). - p
Isn't that exactly what MS and Adobe are trying to do?
Perhaps 60 Mb is only the portion of the code base that resides in our universe. The rest of the O/S could be running on a cluster of quantum computers beyond the event horizon which communicate via a wormhole with the instance on the user's PC. Hence the name. Microsoft's future OSes running in an alternative universe is a win-win as far as I'm concerned. In fact, I'd be thrilled if they moved their HQ there. Can we vote them off the planet? Sorry, sarcasm and ideology aside (and assuming there's no nefarious agenda which there may well be), I think it's great that they've established a group to experiment with new OS concepts. Welcome to the '90's MS.
On the other hand, if there's a market segment used to paying somebody for their software, and paying every year to keep it running, wouldn't that be a great market segment to expand into if you're looking for revenue? This sounds to me like a solid motivation for Zend to better support the Windows world (or at least issue PR statements to this effect). Side note: I'm not suggesting that Zend will get more money out of PHP users on Win directly, but MS will probably get more server rev. if PHP is better supported on Win, which might mean that MS provides some incentive to Zend.
I think every once in awhile some executive on the fringe of the FOSS world gets a hair up his ass and assumes he can monetize some market leading piece of FOSS- completing missing the point that the market share was probably enabled by two things: quality code thats does something useful and available at no/low cost. As soon as you try to siphon off revenue, you're going to upset the applecart and the result may be a decline in market share. Sorry- free rides are pretty hard to come by no matter how clever you are with smoke and mirrors or bait & switch.