Which is the bogus claim, as anyone who's worked with ESXi and developing drivers for it can testify(which I have done). ESX, otoh, ran sort of parallell with the Linux kernel, but ESX was discontinued years ago.
No, the diagram is very simplified, only showing 3 layers, and that the alleged Linux kernel and ESXi hypervisor kernel occupy kernel space. It is very disingenious, since ESXi is bare metal and stand-alone, unlike ESX, which was discontinued years ago. In fact, when reading the link to mailing lists someone in the thread posted, many of the complaints are about ESX, and invalid for ESXi.
In fact, the entire situation smells fishy.... I wonder if they are trying to appropriate VMware code somehow to improve Xen code.
But, ESX is not ESXi. ESXi is very different from ESX(and ESX was discontinued years ago). If they are really suing VMWare over ESXi but based on arguments about ESX, then it only looks like an ideology-based extortion-via-lawsuit attempt.
The entire thing is highly confusing, since ESXi tossed out the Linux kernel components from the hypervisor and runs on bare metal since 4.1(released in july 2010), and ESX, which is not the same thing as ESXi . And BusyBox is not a part of the kernel. Also, the source code for the Open Source components for ESXi 5 have been available for download since the start. So, judging by that technical FAQ, is their argument "You share kernel space with GPL'd stuff, now hand us the goodies or we'll sue you!"?. (Which wouldn't surprise me, given the religious zealotry behind GPL which can easily go into "What's yours is mine, and what's mine makes yours mine too")
Red Hat have made noises in that direction via support channels the last few years, and an argument has been made that their support techs should count as official users on the systems, for GPL reasons. Needless to say, some healthcare IT administrators, defense contractors, police IT etc weren't exactly thrilled by that argument, and quite a few of those have now either swapped to CentOS and do all support in-house, or dumped anything originating from Red Hat and either swap to another distro family or roll their own.
Actually, enough details of Source 2 have been unveiled to start making some judgements. From what I've seen, it's the same usual Valve clusterfuck under the hood, meaning bugs bugs bugs and cheat friendly has hell, a workflow that is convoluted to the point that 8 spiders on LSD and meth in zero-grav can't create a bigger mess....
Actually, you are about a year and a half to two out of date with some of your assertions:
The leadership HAS changed. For example, the CEO was ousted, and the senior legal counsel, responsible for the excessive, almost scientology-like IP protection, went out the hard way back in august 2013. Since then, the climate HAS changed: There have been more licenses for computer games based on their IP, mod makers no longer get instant threatening C&D letters. I've made some tilesets for 40k/Dark Heresy/Rogue Trader/Deathwatch use, and I actually got a very polite mail about 8 months ago, essentially saying they didn't mind, as long as I didn't try to make money from it. A WH fantasy mod for the Total War series has, to my knowledge, not been harassed after the senior legal counsel was fired, despite Creative Assembly having received a license for Warhammer Fantasy.
Right now, GW ARE changing: For example, the new Warhammer Fantasy that is in the works is apparently going back to older roots, with less hard restrictions on units, instead going back to the old percentage allocations(At least x% points on troops, at most y% points on characters, that kind of thing), as well as going towards more skirmish type(I wonder if they finally dumped Nigel fucking Stillmann, the chief asshole behind the move to put as many and as large units as possible on the table... A bit like how Gav Thorpe and Ian Pickstock helped fuck up WH40k 3rd ed and onwards...)
Also, in terms of models, what it looks like, according to some people who are in the pipeline, is that GW are changing focus towards Forge World, and possibly even customized miniatures, and possibly just selling cad files etc for basic troops.
Wind resistance, real slopes, fighting slippery surfaces, absorbing shocks and bumps, changing weather conditions, maintaining your concentration to a far sharper degree than a sim demands all cause you to burn more energy than any simulator you use.
It's why you see so many gym stars perform like shit when facing the real thing. Or why iRacing(a pretty good racing simulator) stars, who outperform real racing drivers in the sim, are like 20 seconds behind the same racing drivers when placed in a real racing car, even after 2 weeks of training in the real car.
"- If there's a problem,
- your own developpers will be pulling out their hair because it's not compliant to their way of working (or worse unreadable spaghetti code)
- no garantees, you've already validated and payed the freelancer, yer on yer own."
I feel that I have to comment here, from my experience.
As part of my contracts, I keep specifically to their code templates, variable naming etc, as long as it doesn't affect reliability/security/safety, and that's also laid out in the terms of the contract.
In terms of guarantees, many of us specialists who don't work on run-of-the-mill contracts have support contracts attached to our work.
Indeed, one of the benefits us freelancers can bring to a project is the lack of internal politics and development baggage, as well as experience from other projects, which can counteract the drawbacks of not being as in-the-loop as the long-term employees.
Also, if you're a specialist, keep in mind that that's the reason you're brought in, in case the long-term employees get arrogant and dismissive: You're a specialist on something they need, and they don't have the knowledge or experience on that in-house.
Serious freelancers who try to make a decent living out of it don't use those sites, for several reasons, some of which I'll go through.
1: Overall(note the word, overall), they cater to simple projects in oversaturated fields.
2: They get flooded by unscrupulous or simply cheap people who offer pay way below what is decent.
3: You get no way of building up a decent reputation.
4: As a combination of the above factors, you have to churn through lots of contracts constantly, increasing risk of burnout, failed contracts etc.
On the other hand, to make a decent living, both in pay and in the way of hours you work, you want to work in specialist niches, one contract at a time, maybe two overlapping at a pinch, if they don't interfere with each other(Starting the design phase of a new project as you're working on the testing/debugging/deployment phase of your previous one works ok usually, while starting a new project while in the development/coding phase of the previous one is usually not so good...)
You want to establish a good reputation and a wide contact network. And always make sure that you have a lawyer of your own go through the contracts, and the help of a good accountant. In fact, the more familiarity you get with your clients, the better, since you will get more leeway in case of sickness/family issues/issues beyond your control etc.
Avoiding the use of those websites, and working via agents instead, also gives you better options for negotiation(especially if you have the advise of a lawyer and/or an accountant, depending on the issues you need advise on), and can better structure your life. It also allows you to check out potential clients much more easily. Some of your contacts, or your contacts contacts, may know about some issues that have not made it into public records for example. Point in case, a contract was offered to my agent once, which he immediately blacklisted. Why? Because he checked up on some of the people running the company, and found major financial discrepancies, such as the company nominally running at a loss, CEO supposedly earning only 20k euro per year from that post and a total yearly income of 40k euro per year yet still owning a yacht worth about 2M euro etc.
I have freelanced for about 15 years now, and while I initially had to take risks with many contracts, I can now be far more careful, and choose the contract offers that will benefit me not just financially, but also what best suits my health and family.
Now, by no means do I earn any extreme amounts. Last fiscal year, I earned about 50k euro after taxes, which may not seem like much, but in terms of swedish living costs, that's well above average. However, as a freelancer, I do have to set money aside for courses, seminars etc.
(OK, I'll take that question from the clueless nerd in the peanut gallery)
Why do I set aside money for courses, seminars etc, when I could just use google and study on my own?
Well, as I pointed out above, contacts and reputation are everything if you want to be successful, and don't want to be screwed over. To the intelligent AND wise people, it also means exchange of experiences, those things you can't teach via text tutorials etc. It means getting in touch with new people who can forward things your way, as you forward things to them. In terms of reputation, one of the way it helps is that participating in courses etc lets you be seen as still keeping in touch, still able to learn, that you are not stagnating and too heavily wedged inside a niche.
It only partially helps people. You also have to find out other factors in someones daily life that contribute to malnutrition of any kind. As one of the other posters mentioned, if you blindly follow your simplistic notion, there are no compensations made for stress, illness, symptoms of lack of nutrition etc. As an example, during a hospital treatment I was in for, a dietician had me on a strict calorie plan, ignoring that I was doing physical therapy, and was healing injuries, and thus almost caused me serious complications. I did begin to suffer starvation hallucinations, bouts of depression etc.
Not that Akamai's State of the Internet is worth a damn anyway, with the throttled shit we have to deal with in the nordic countries. Seriously, Akamai is crap here. Steam, Limelight Networks etc etc, I can max out my 100/100 connection. If it's Akamai, it slows down to like 20Mbit/s.
Personally, I like the quieter cars, both in everyday traffic and in racing. Unlike many others, I enjoy the new turbo V6's in Formula One for example, and it would be interesting to see how much faster the turbo V6's would be than the previous eras if they were allowed to use the aerodynamics regs of those eras(That's what actually slowed down the 2014 F1 cars, the greater restrictions on aero).
I also enjoy the LMP1 hybrids that are much quieter than their spiritual ancestors, the Group C prototypes.
For me, within a given engine type, more noise=less impressive, since it shows that it's badly engineered and wasting energy.
6 WD Red 2TB disks split over 3 VDEVs, sector size is set correctly etc. No encryption, no compression. And it makes no difference whether I use NFS or Samba.
ZFS is something with which I have yet to familiarize myself with the internals so I can only guess, but my initial impression is that it's similar to older unix filesystems(and why Silicon Graphics developed XFS) in that it is not that good at handling many large files simultaneously. So I have the original video clip, then I have individual folders with the RBG channel images, the alpha channel images, the shadow maps, etc etc etc, meaning that for each second of 3D animation there's hundreds of images.
I've recently gone back to my roots and started dabbling with 3D animation and compositing again. My fileserver is a FreeBSD machine running on a decent 64-bit CPU with 16GiB RAM, with ZFS. And let me tell you, ZFS is dog slow for some uses, without it being anywhere near full. In my case, lossless-encoded video, and directories with thousands of 4MiB+ images, and working against that in realtime(or trying to), the filesystem stalled out at 80MiB/s, while my old fileserver running Linux and XFS easily saturated the gigabit link
The mainframe people I know, when they rarely refer to transactions, have a slightly different meaning from when windows or unix people do it. The mainframe people more often rever to messages, which is a whole discrete task, which can often require multiple database transactions, some computational passes etc. They usually talk about hundreds of thousands of messages per hour, so if it's 2.5 billion mainframe-style "transactions"(messages), it's pretty damn impressive.
Let's not forget something as simple as debugging. Someone who understands the hardware and knows Assembler will always have an advantage when it comes to debugging.
Hardware should always be a concern, because hardware is the reality that implements the abstraction of a program. No matter how efficient something is in purely mathematical terms, it's the hardware that determines the actual performance, complexity and problems. ISA, I/O capabilities, amount of RAM etc all matter in deciding what will be the best way to implement something.
No matter how many layers of abstraction you put in to provide the illusion of being able to ignore the hardware, the reality of hardware will always matter.
Nono, like other big IT projects in the UK, it will be using "the very latest in Agile know-how", and cost 3 times as much as any clusterfuck that involves Oracle, take 50% longer, and spread 300% more blame on "old fossiles"....
Disclaimer: Had to interface with a EU project under UK IT auspices last year.... Painful....
Slashdot Mirror
Which is the bogus claim, as anyone who's worked with ESXi and developing drivers for it can testify(which I have done). ESX, otoh, ran sort of parallell with the Linux kernel, but ESX was discontinued years ago.
But with ESXi, the linux kernel runs on top of the hypervisor.
No, the diagram is very simplified, only showing 3 layers, and that the alleged Linux kernel and ESXi hypervisor kernel occupy kernel space. It is very disingenious, since ESXi is bare metal and stand-alone, unlike ESX, which was discontinued years ago. In fact, when reading the link to mailing lists someone in the thread posted, many of the complaints are about ESX, and invalid for ESXi.
In fact, the entire situation smells fishy.... I wonder if they are trying to appropriate VMware code somehow to improve Xen code.
And GPL is very much "what's yours is mine and what's mine also makes yours mine"
But, ESX is not ESXi. ESXi is very different from ESX(and ESX was discontinued years ago). If they are really suing VMWare over ESXi but based on arguments about ESX, then it only looks like an ideology-based extortion-via-lawsuit attempt.
The entire thing is highly confusing, since ESXi tossed out the Linux kernel components from the hypervisor and runs on bare metal since 4.1(released in july 2010), and ESX, which is not the same thing as ESXi . And BusyBox is not a part of the kernel. Also, the source code for the Open Source components for ESXi 5 have been available for download since the start. So, judging by that technical FAQ, is their argument "You share kernel space with GPL'd stuff, now hand us the goodies or we'll sue you!"?. (Which wouldn't surprise me, given the religious zealotry behind GPL which can easily go into "What's yours is mine, and what's mine makes yours mine too")
Red Hat have made noises in that direction via support channels the last few years, and an argument has been made that their support techs should count as official users on the systems, for GPL reasons. Needless to say, some healthcare IT administrators, defense contractors, police IT etc weren't exactly thrilled by that argument, and quite a few of those have now either swapped to CentOS and do all support in-house, or dumped anything originating from Red Hat and either swap to another distro family or roll their own.
Actually, that's exactly what I had in mind :p
(and you forgot to consider the fact that the spiders would be on both meth and LSD, and there's 8 of the spiders at the same time :p )
Source 2 will work.... But working with it looks like it'll only be a slightly less painful experience than Source
Actually, enough details of Source 2 have been unveiled to start making some judgements. From what I've seen, it's the same usual Valve clusterfuck under the hood, meaning bugs bugs bugs and cheat friendly has hell, a workflow that is convoluted to the point that 8 spiders on LSD and meth in zero-grav can't create a bigger mess....
Actually, you are about a year and a half to two out of date with some of your assertions:
The leadership HAS changed. For example, the CEO was ousted, and the senior legal counsel, responsible for the excessive, almost scientology-like IP protection, went out the hard way back in august 2013. Since then, the climate HAS changed: There have been more licenses for computer games based on their IP, mod makers no longer get instant threatening C&D letters. I've made some tilesets for 40k/Dark Heresy/Rogue Trader/Deathwatch use, and I actually got a very polite mail about 8 months ago, essentially saying they didn't mind, as long as I didn't try to make money from it. A WH fantasy mod for the Total War series has, to my knowledge, not been harassed after the senior legal counsel was fired, despite Creative Assembly having received a license for Warhammer Fantasy.
Right now, GW ARE changing: For example, the new Warhammer Fantasy that is in the works is apparently going back to older roots, with less hard restrictions on units, instead going back to the old percentage allocations(At least x% points on troops, at most y% points on characters, that kind of thing), as well as going towards more skirmish type(I wonder if they finally dumped Nigel fucking Stillmann, the chief asshole behind the move to put as many and as large units as possible on the table... A bit like how Gav Thorpe and Ian Pickstock helped fuck up WH40k 3rd ed and onwards...)
Also, in terms of models, what it looks like, according to some people who are in the pipeline, is that GW are changing focus towards Forge World, and possibly even customized miniatures, and possibly just selling cad files etc for basic troops.
Wind resistance, real slopes, fighting slippery surfaces, absorbing shocks and bumps, changing weather conditions, maintaining your concentration to a far sharper degree than a sim demands all cause you to burn more energy than any simulator you use.
It's why you see so many gym stars perform like shit when facing the real thing. Or why iRacing(a pretty good racing simulator) stars, who outperform real racing drivers in the sim, are like 20 seconds behind the same racing drivers when placed in a real racing car, even after 2 weeks of training in the real car.
I don't know, it could be worse.... There could have been sizeable amounts of elemental fluorine and chlorine too..... And other very nasty things...
"- If there's a problem,
- your own developpers will be pulling out their hair because it's not compliant to their way of working (or worse unreadable spaghetti code)
- no garantees, you've already validated and payed the freelancer, yer on yer own."
I feel that I have to comment here, from my experience.
As part of my contracts, I keep specifically to their code templates, variable naming etc, as long as it doesn't affect reliability/security/safety, and that's also laid out in the terms of the contract.
In terms of guarantees, many of us specialists who don't work on run-of-the-mill contracts have support contracts attached to our work.
Indeed, one of the benefits us freelancers can bring to a project is the lack of internal politics and development baggage, as well as experience from other projects, which can counteract the drawbacks of not being as in-the-loop as the long-term employees.
Also, if you're a specialist, keep in mind that that's the reason you're brought in, in case the long-term employees get arrogant and dismissive: You're a specialist on something they need, and they don't have the knowledge or experience on that in-house.
Serious freelancers who try to make a decent living out of it don't use those sites, for several reasons, some of which I'll go through.
1: Overall(note the word, overall), they cater to simple projects in oversaturated fields.
2: They get flooded by unscrupulous or simply cheap people who offer pay way below what is decent.
3: You get no way of building up a decent reputation.
4: As a combination of the above factors, you have to churn through lots of contracts constantly, increasing risk of burnout, failed contracts etc.
On the other hand, to make a decent living, both in pay and in the way of hours you work, you want to work in specialist niches, one contract at a time, maybe two overlapping at a pinch, if they don't interfere with each other(Starting the design phase of a new project as you're working on the testing/debugging/deployment phase of your previous one works ok usually, while starting a new project while in the development/coding phase of the previous one is usually not so good...)
You want to establish a good reputation and a wide contact network. And always make sure that you have a lawyer of your own go through the contracts, and the help of a good accountant. In fact, the more familiarity you get with your clients, the better, since you will get more leeway in case of sickness/family issues/issues beyond your control etc.
Avoiding the use of those websites, and working via agents instead, also gives you better options for negotiation(especially if you have the advise of a lawyer and/or an accountant, depending on the issues you need advise on), and can better structure your life. It also allows you to check out potential clients much more easily. Some of your contacts, or your contacts contacts, may know about some issues that have not made it into public records for example. Point in case, a contract was offered to my agent once, which he immediately blacklisted. Why? Because he checked up on some of the people running the company, and found major financial discrepancies, such as the company nominally running at a loss, CEO supposedly earning only 20k euro per year from that post and a total yearly income of 40k euro per year yet still owning a yacht worth about 2M euro etc.
I have freelanced for about 15 years now, and while I initially had to take risks with many contracts, I can now be far more careful, and choose the contract offers that will benefit me not just financially, but also what best suits my health and family.
Now, by no means do I earn any extreme amounts. Last fiscal year, I earned about 50k euro after taxes, which may not seem like much, but in terms of swedish living costs, that's well above average. However, as a freelancer, I do have to set money aside for courses, seminars etc.
(OK, I'll take that question from the clueless nerd in the peanut gallery)
Why do I set aside money for courses, seminars etc, when I could just use google and study on my own?
Well, as I pointed out above, contacts and reputation are everything if you want to be successful, and don't want to be screwed over. To the intelligent AND wise people, it also means exchange of experiences, those things you can't teach via text tutorials etc. It means getting in touch with new people who can forward things your way, as you forward things to them. In terms of reputation, one of the way it helps is that participating in courses etc lets you be seen as still keeping in touch, still able to learn, that you are not stagnating and too heavily wedged inside a niche.
It only partially helps people. You also have to find out other factors in someones daily life that contribute to malnutrition of any kind. As one of the other posters mentioned, if you blindly follow your simplistic notion, there are no compensations made for stress, illness, symptoms of lack of nutrition etc. As an example, during a hospital treatment I was in for, a dietician had me on a strict calorie plan, ignoring that I was doing physical therapy, and was healing injuries, and thus almost caused me serious complications. I did begin to suffer starvation hallucinations, bouts of depression etc.
Since we're in the EU, as is Hungary, in fact, yes, we do!
Not that Akamai's State of the Internet is worth a damn anyway, with the throttled shit we have to deal with in the nordic countries. Seriously, Akamai is crap here. Steam, Limelight Networks etc etc, I can max out my 100/100 connection. If it's Akamai, it slows down to like 20Mbit/s.
Personally, I like the quieter cars, both in everyday traffic and in racing. Unlike many others, I enjoy the new turbo V6's in Formula One for example, and it would be interesting to see how much faster the turbo V6's would be than the previous eras if they were allowed to use the aerodynamics regs of those eras(That's what actually slowed down the 2014 F1 cars, the greater restrictions on aero).
I also enjoy the LMP1 hybrids that are much quieter than their spiritual ancestors, the Group C prototypes.
For me, within a given engine type, more noise=less impressive, since it shows that it's badly engineered and wasting energy.
There is a way to work around that. Create a VHD on your network drive, mount that VHD as a drive via disk manager, point backup to that drive.
6 WD Red 2TB disks split over 3 VDEVs, sector size is set correctly etc. No encryption, no compression. And it makes no difference whether I use NFS or Samba.
ZFS is something with which I have yet to familiarize myself with the internals so I can only guess, but my initial impression is that it's similar to older unix filesystems(and why Silicon Graphics developed XFS) in that it is not that good at handling many large files simultaneously. So I have the original video clip, then I have individual folders with the RBG channel images, the alpha channel images, the shadow maps, etc etc etc, meaning that for each second of 3D animation there's hundreds of images.
I've recently gone back to my roots and started dabbling with 3D animation and compositing again. My fileserver is a FreeBSD machine running on a decent 64-bit CPU with 16GiB RAM, with ZFS. And let me tell you, ZFS is dog slow for some uses, without it being anywhere near full. In my case, lossless-encoded video, and directories with thousands of 4MiB+ images, and working against that in realtime(or trying to), the filesystem stalled out at 80MiB/s, while my old fileserver running Linux and XFS easily saturated the gigabit link
The mainframe people I know, when they rarely refer to transactions, have a slightly different meaning from when windows or unix people do it. The mainframe people more often rever to messages, which is a whole discrete task, which can often require multiple database transactions, some computational passes etc. They usually talk about hundreds of thousands of messages per hour, so if it's 2.5 billion mainframe-style "transactions"(messages), it's pretty damn impressive.
Let's not forget something as simple as debugging. Someone who understands the hardware and knows Assembler will always have an advantage when it comes to debugging.
Hardware should always be a concern, because hardware is the reality that implements the abstraction of a program. No matter how efficient something is in purely mathematical terms, it's the hardware that determines the actual performance, complexity and problems. ISA, I/O capabilities, amount of RAM etc all matter in deciding what will be the best way to implement something.
No matter how many layers of abstraction you put in to provide the illusion of being able to ignore the hardware, the reality of hardware will always matter.
Nono, like other big IT projects in the UK, it will be using "the very latest in Agile know-how", and cost 3 times as much as any clusterfuck that involves Oracle, take 50% longer, and spread 300% more blame on "old fossiles"....
Disclaimer: Had to interface with a EU project under UK IT auspices last year.... Painful....