>Have you taken any steps to confirm this isn't a localised problem on your system? I tried, but whenever I have taken the time to confirm it, I'm two versions behind and they don't accept the bug report.
>Let's just write a new list, like "can program a VCR" and start over. Great! 80% of humanity will be dropped into the "sub-human" category and be available as cheap slaves! Perhaps even my boss!
.... anyway, what the hell do they change from version to version?
If they tell you "Changes are not *dangerous*, because we stick to standards", then that is bullshit. If a change is "not at all dangerous" then it is also "not at all necessary", since it would imply the change does not change anything. What I have seen in 15 years in IT is that even some pretty minor thing that changed in a software product can bring your work flow to a halt. And you can lose business for hours or days.
They can probably "set up clouds" in a few hours. You need connectivity and storage. They probably already have connectivity, and they already sell storage products.
The only thing they needed was to install a few of their off-the-shelf SANs with their off-the-shelf clustering solutions in a few of their already existing data centres. The MOST work would probably have been to put software interfaces on their devices that talk to that storage. But that's also pretty standard stuff.
I wonder how many executives will curse the day they fired a "IT guy that made them feel bad" and replaced him with a "Cloud that ignores them" in case of problems.
But I can do pretty much NOTHING for YOU until you tell me what YOU want.
I might program something great for ME on my own, which might just by a freak coincidence be also great for YOU, but that's not working in the real world 99% of the time.
I have seen the tendency in computing over the last couple of years. It's no longer "OK, I have to solve problem X. What tool can I use to solve it?" it's "Oh, I have Shiny new tool X. what Problems could I solve with it?"
If people did that in the real world, you would have tons of people trying to fly across the Atlantic on rubber ducks, while taking a brand new Airbus to the pool as a flotation device.
For one, your tail lights are not bright enough for the bigger distances, so a Vogon construction cruiser will probably plough right into you when you are stuck at the end of a traffic jam.
... these days is that even *BEFORE* you get accustomed to something (never mind restructuring your entire work flow and setup, like in this case not being able to put the temporary scratch - files in a temporary place, which would require you to change your storage setup), they will change it again.
Also, when you "have to spend time" with a new version before you can use it, you can NEVER install that new version while you are in the middle of a project with deadlines. Which is, like, pretty much all of the time, when you are good at what you do.
Yea. Just install Version 5. On a few hundred provisioned images and thousands of machines. Where it took 3 years to completely switch from IE6 to IE7 with all internal web applications. In an environment where People are still pissed by the UI changes from FF3.6 to FF4.0.
That's like "I don't get what the hell is wrong with this pile of shit on the chocolate cake.. Just f'in take it off! Problem solved!"
That's why I need to know what is an update and what is a new version. Until now I just needed to look at the version number, now I have to *investigate* and *waste time* to find out if the new version is really a new version or not.
That's why I hate the new scheme. The logic should be, from my point as a developer:
- Minor version numbers don't change interfaces - Major version numbers might change interfaces.
From now on, how will they alert people that the change from version 4 to 5 to 6 doesn't change interfaces, but the one from 6 to 7 does?
One of the main reasons I liked OSS better than Commercial applications was that they put more focus on development than on marketing. But that seems to change. Crap. Perhaps, similar to the Mozilla -> Phoenix fork back then, that aimed to cut the bloat, there now could be a need for a fork that cuts the marketing.
That's even faster. Connect disc to controller, do a copy of the raw device.
One that *will* take time is to come up with a system of checks and procedures that the chain of evidence is proven. But that has to be done BEFORE the process is started. So it has no bearing on the time the process itself takes.
The other thing that *will* take time is to figure out what the data mean, if and how it was RAIDed (pun intended) or encrypted. But that also has no effect on how long it takes to make a 1:1 copy of the bits.
"Hundreds of Megabytes"? Like, the size of a CD? Takes a few seconds on my end here.
As for the "Terrabytes", here is the end of the log from one of or tape backups, to a 6 year old tape library:
Total number of bytes transferred: 247.79 GB Data transfer time: 3,595.75 sec Network data transfer rate: 72,261.73 KB/sec Aggregate data transfer rate: 53,638.98 KB/sec Objects compressed by: 0% Elapsed processing time: 01:20:44
A quarter of a TB in 1:20 h, so a copying an entire Terabyte to tape over a 1Gbps Network could probably be done in about 5 hours or so.
So they basically just *bundle* the "voice plug-in" into the standard install.
What I'm really waiting for is for the "to the same stuff you have been doing for years, only with added layers of complexity and corporate limitations" bubble to burst.
It used to be that you could run one application at a time on your computer. Then came the multitasking-OS, and the promise of "run as much applications in parallel as you like". Then the applications became difficult to maintain, and conflicted with each other, and they started to put different OS installs into different Virtual machines. Then maintaining the applications became even more of a hassle, and they start putting them into the browser.
Of course, as an Administrator for a medium sized company, I now have to have VMs with DIFFERENT VERSIONS OF DIFFERENT BROWSERS handy, because a lot of those co-called "easy deployable web-apps" only work with specific browsers often only with specific versions of those browsers.
Plus the ADDITIONAL headache that possibly confidential data is accessed with a web browser that is also used to browse the internet. Since you can basically never be sure someone might find a vulnerability in a browser that can be exploited, you have to someone SPLIT the "Browser that is installed for real web browsing" and the "Browser that is installed to access you internal web interfaces" somehow again.
The way I see it the whole HTML5 craze is just re-inventing the UI (Windows/GTK/QT/etc..) again badly. It might work for "Fun Leisure Gadgets", but I wonder if there is anybody still developing for the people who really have to WORK with their computers. In that way the age of the "Personal Computer" might come to an end. 95% of people will just use "Smart" Phones and "Smart" TV to do a little browsing and mailing, only very few will still have and use a real full-featured computer. Which might be bad news price-wise for those.
It would *technically* make more sense to have.comp and.music TLDs with the brand beneath them.
Of course all of that is pretty much irrelevant for a *users* standpoint. There it could make more sense to split the "domain" system into a keyword system including brand / field / country / product for example, where keywords can be registered. Order basically doesn't matter, what each field MEANS matters.
The only thing I see in this ICANN proposal here is a money grab my ICANN.
I have actually read a few articles about that phenomenon. There seem to be some developments in the work to alleviate the Problem.
For example, there is a technology called "RAID" (Real-time Automatic Information Duster) that keeps the information free of aggregating dust and other particles, like network sewage and electric soot, which can be quite a problem in machines with faster processors. The Linux version works quite well. Unfortunately the Windows version seems to be available only with with an automatic indexing product that slows the machine down pretty bad. (called Windex, I believe). And it's not available at all for those Mac things. There the proprietary there iBlow software seems to keep the data on the mac clean all right, but it blows all the dirt to adjacent non-Apple machines on the network. That's why it is so important to have your dustwall up to date.
They should just release when they have something to release.
What's the point of releasing version after version when the original version was working and everybody using it was happy?
I can understand commercial software trying to make money, but for FOSS they could just as well leave the version that people want in place until there is a demand for something new in the community, and they have thought things through collectively.
So instead of a real "start menu" you get a search box, perhaps with tab auto-completion of commands?
Kinda, like, you know, a command line,only with less features? So they finally realized that typing a command is way faster than hunting for it in a menu?;-P
Slashdot Mirror
>Have you taken any steps to confirm this isn't a localised problem on your system?
I tried, but whenever I have taken the time to confirm it, I'm two versions behind and they don't accept the bug report.
>Let's just write a new list, like "can program a VCR" and start over.
Great! 80% of humanity will be dropped into the "sub-human" category and be available as cheap slaves! Perhaps even my boss!
.... anyway, what the hell do they change from version to version?
If they tell you "Changes are not *dangerous*, because we stick to standards", then that is bullshit. If a change is "not at all dangerous" then it is also "not at all necessary", since it would imply the change does not change anything. What I have seen in 15 years in IT is that even some pretty minor thing that changed in a software product can bring your work flow to a halt. And you can lose business for hours or days.
They can probably "set up clouds" in a few hours. You need connectivity and storage. They probably already have connectivity, and they already sell storage products.
The only thing they needed was to install a few of their off-the-shelf SANs with their off-the-shelf clustering solutions in a few of their already existing data centres.
The MOST work would probably have been to put software interfaces on their devices that talk to that storage. But that's also pretty standard stuff.
Well, take the visual out of data VISUALIZATION, what is left?
Yes. Then you are a trained CAD tracer and a computer geek. Not a trained engineer.
Simulations can supplement the training. But just like in the military or for pilots, simulation alone is not enough.
That's the problem these days.
If humans had ever thought that way, we'd still be living in caves. Very few people who ever started something like a cathedral ever saw it finished.
I wonder how many executives will curse the day they fired a "IT guy that made them feel bad" and replaced him with a "Cloud that ignores them" in case of problems.
Amen to that Brother. And I *AM* a programmer.
But I can do pretty much NOTHING for YOU until you tell me what YOU want.
I might program something great for ME on my own, which might just by a freak coincidence be also great for YOU, but that's not working in the real world 99% of the time.
I actually wondered during lunch break if it is possible to get IE8 to run under Wine.
So far I just decided to download the 4.01 Firefox source tree for safekeeping and disable all auto-updates.
I have seen the tendency in computing over the last couple of years. It's no longer "OK, I have to solve problem X. What tool can I use to solve it?" it's "Oh, I have Shiny new tool X. what Problems could I solve with it?"
If people did that in the real world, you would have tons of people trying to fly across the Atlantic on rubber ducks, while taking a brand new Airbus to the pool as a flotation device.
For one, your tail lights are not bright enough for the bigger distances, so a Vogon construction cruiser will probably plough right into you when you are stuck at the end of a traffic jam.
... these days is that even *BEFORE* you get accustomed to something (never mind restructuring your entire work flow and setup, like in this case not being able to put the temporary scratch - files in a temporary place, which would require you to change your storage setup), they will change it again.
Also, when you "have to spend time" with a new version before you can use it, you can NEVER install that new version while you are in the middle of a project with deadlines. Which is, like, pretty much all of the time, when you are good at what you do.
Yea. Just install Version 5. On a few hundred provisioned images and thousands of machines. Where it took 3 years to completely switch from IE6 to IE7 with all internal web applications. In an environment where People are still pissed by the UI changes from FF3.6 to FF4.0.
That's like "I don't get what the hell is wrong with this pile of shit on the chocolate cake .. Just f'in take it off! Problem solved!"
That's why I need to know what is an update and what is a new version. Until now I just needed to look at the version number, now I have to *investigate* and *waste time* to find out if the new version is really a new version or not.
That's why I hate the new scheme. The logic should be, from my point as a developer:
- Minor version numbers don't change interfaces
- Major version numbers might change interfaces.
From now on, how will they alert people that the change from version 4 to 5 to 6 doesn't change interfaces, but the one from 6 to 7 does?
One of the main reasons I liked OSS better than Commercial applications was that they put more focus on development than on marketing. But that seems to change. Crap. Perhaps, similar to the Mozilla -> Phoenix fork back then, that aimed to cut the bloat, there now could be a need for a fork that cuts the marketing.
That's even faster. Connect disc to controller, do a copy of the raw device.
One that *will* take time is to come up with a system of checks and procedures that the chain of evidence is proven. But that has to be done BEFORE the process is started. So it has no bearing on the time the process itself takes.
The other thing that *will* take time is to figure out what the data mean, if and how it was RAIDed (pun intended) or encrypted. But that also has no effect on how long it takes to make a 1:1 copy of the bits.
"Hundreds of Megabytes"? Like, the size of a CD? Takes a few seconds on my end here.
As for the "Terrabytes", here is the end of the log from one of or tape backups, to a 6 year old tape library:
Total number of bytes transferred: 247.79 GB
Data transfer time: 3,595.75 sec
Network data transfer rate: 72,261.73 KB/sec
Aggregate data transfer rate: 53,638.98 KB/sec
Objects compressed by: 0%
Elapsed processing time: 01:20:44
A quarter of a TB in 1:20 h, so a copying an entire Terabyte to tape over a 1Gbps Network could probably be done in about 5 hours or so.
So they basically just *bundle* the "voice plug-in" into the standard install.
What I'm really waiting for is for the "to the same stuff you have been doing for years, only with added layers of complexity and corporate limitations" bubble to burst.
It used to be that you could run one application at a time on your computer. Then came the multitasking-OS, and the promise of "run as much applications in parallel as you like". Then the applications became difficult to maintain, and conflicted with each other, and they started to put different OS installs into different Virtual machines. Then maintaining the applications became even more of a hassle, and they start putting them into the browser.
Of course, as an Administrator for a medium sized company, I now have to have VMs with DIFFERENT VERSIONS OF DIFFERENT BROWSERS handy, because a lot of those co-called "easy deployable web-apps" only work with specific browsers often only with specific versions of those browsers.
Plus the ADDITIONAL headache that possibly confidential data is accessed with a web browser that is also used to browse the internet. Since you can basically never be sure someone might find a vulnerability in a browser that can be exploited, you have to someone SPLIT the "Browser that is installed for real web browsing" and the "Browser that is installed to access you internal web interfaces" somehow again.
The way I see it the whole HTML5 craze is just re-inventing the UI (Windows/GTK/QT/etc..) again badly. It might work for "Fun Leisure Gadgets", but I wonder if there is anybody still developing for the people who really have to WORK with their computers. In that way the age of the "Personal Computer" might come to an end. 95% of people will just use "Smart" Phones and "Smart" TV to do a little browsing and mailing, only very few will still have and use a real full-featured computer. Which might be bad news price-wise for those.
Yep. *THEY* don't have to deal with us, *WE* don't have to deal with them. ;-P
And who pays for the .apple tld then?
It would *technically* make more sense to have .comp and .music TLDs with the brand beneath them.
Of course all of that is pretty much irrelevant for a *users* standpoint. There it could make more sense to split the "domain" system into a keyword system including brand / field / country / product for example, where keywords can be registered. Order basically doesn't matter, what each field MEANS matters.
The only thing I see in this ICANN proposal here is a money grab my ICANN.
Kids these days. We had hot snow and freezing ash all day long. I seriously clogged up the punch cards.
Did we sue? NO, we just made do with what we had.
I have actually read a few articles about that phenomenon. There seem to be some developments in the work to alleviate the Problem.
For example, there is a technology called "RAID" (Real-time Automatic Information Duster) that keeps the information free of aggregating dust and other particles, like network sewage and electric soot, which can be quite a problem in machines with faster processors. The Linux version works quite well. Unfortunately the Windows version seems to be available only with with an automatic indexing product that slows the machine down pretty bad. (called Windex, I believe). And it's not available at all for those Mac things. There the proprietary there iBlow software seems to keep the data on the mac clean all right, but it blows all the dirt to adjacent non-Apple machines on the network. That's why it is so important to have your dustwall up to date.
They should just release when they have something to release.
What's the point of releasing version after version when the original version was working and everybody using it was happy?
I can understand commercial software trying to make money, but for FOSS they could just as well leave the version that people want in place until there is a demand for something new in the community, and they have thought things through collectively.
So instead of a real "start menu" you get a search box, perhaps with tab auto-completion of commands?
Kinda, like, you know, a command line,only with less features? So they finally realized that typing a command is way faster than hunting for it in a menu? ;-P