OSDev.org

Brendan wrote:If pretty positive that, if my OS is ever usable/released, users will look back at the whining and criticism I get from various people (who won't even consider doing anything more than poor re-implementations of tired old ideas), and be completely unable to understand how I manage to resist the temptation to unleash the "wrathful purge" that this forum desperately requires in order to reverse its continuing de-evolution towards a "kindergarten idiocracy" (where those that have learnt not to think and to only regurgitate what they've seen elsewhere have somehow managed to decide they are "superior"; despite the fact that even the most inexperienced beginner has far more potential than they do, simply because an inexperience beginner hasn't been infected by the plague of "failure to consider anything different" yet).

Schol-R-LEA wrote:

Brendan wrote:If pretty positive that, if my OS is ever usable/released, users will look back at the whining and criticism I get from various people (who won't even consider doing anything more than poor re-implementations of tired old ideas), and be completely unable to understand how I manage to resist the temptation to unleash the "wrathful purge" that this forum desperately requires in order to reverse its continuing de-evolution towards a "kindergarten idiocracy" (where those that have learnt not to think and to only regurgitate what they've seen elsewhere have somehow managed to decide they are "superior"; despite the fact that even the most inexperienced beginner has far more potential than they do, simply because an inexperience beginner hasn't been infected by the plague of "failure to consider anything different" yet).

Permit me to be blunt: you are talking as if you have never actually worked in professional programming. I assume that this is not true, given that you have been with this group for over a decade and clearly are quite knowledgeable about computer programming, but you give the impression of someone who thinks that computer programming is about designing and writing computer programs, someone who has never seen his special snowflakes get thrown into a big melting pile which some cat then urinates on (to use Peter Welch's oddly apt metaphor).

Nothing could be further from the truth. Those are the least important, and least difficult, parts of software development. The hard parts are all in interpersonal communication, and I have yet to see a mathematically solid algorithm or heuristic to solve that, never mind one to solve things like being hired by an idiot whose sole point of reference in IT is a fifteen-year-old article in Wired that he read while in the dentist's office, or to connect two 'standard-compliant' libraries which not only aren't compliant but are completely unusable, which the boss has mandated the use of come hell or high water.

I have seen companies hire two dozen programmers at rates of between $60K and $120K USD per annum and then keep them idle for six months waiting for programs to be fixed which never came (I was in the group for four of those months, and it was the easiest $20K I ever made, thank you, SIAC and Kaiser Permanente!).

I have seen a company president announce that no testing will be done on the company's software before it ships (and then, when I left, tell the graphic artist, my friend Sophianna, that she needed to learn Access 97 programming ASAP to cover for my absence).

I have seen managers fire programmers for telling the truth, and managers fire programmers for refusing to lie to customers.

I have seen a company president fire my friend Sophie (the same from the earlier story, who had been hired as a web artist and then dragooned into system admin) because she didn't back up a mission-critical drive which he'd refused to buy backup media for just one day prior, backup media which she had been asking him to authorize the purchase of for the previous six months.

I have seen a major bank put every one of their web developers through a $25K-per-head training course only to fire the whole team because they decided to hire offshore experts in a completely different system instead (a move which just happened to coincide with the dropping of the Dot-Bomb, but surely that had nothing to do with it of course...).

And oh yes, I have seen companies - and government offices, and NPOs - where every computer in the place was at least 12 years old, and where there were no prospects for replacing them in the near future.

These aren't edge cases. These are not flukes. And while some were short-lived startups that died of their endemic stupidities, others were big, successful corporations, or major government offices.

This is exactly how the majority of companies treat IT, because even at major software shops, programmers, IT techs, and system administrators are all generally seen as over-priced craftsmen whom they would replace as soon as something that looks cheaper comes along. The majority of companies, even companies whose primary products are software or IT services, view IT as overhead, not capital. Regardless of whether this is right or wrong, it is the truth.

Think I am wrong? If I were, things like The Daily WTF and the Tales From Tech Support subreddit wouldn't exist. The now defunct F***ed Company website wouldn't have existed. The Dot-Com bubble wouldn't have happened at all, because that was driven by people looking, not to make a product, but a quick buck, people who had no idea what 'the information superhighway' was but knew that it was raining venture capital thanks to it.

I would have expected you to have seen things at least as bad as these if you had been an IT professional for even a fraction of the time you have been on this forum. I can't see how you have kept up your belief that you can force people to accept your software solely on the basis of it being "better", according to you, unless you have decided to blame the cat for piddling on your snowflake, instead of either the guys who asked for snowflakes when they really needed soap flakes, or the ones who decided to throw them all into a big pile out under the summer sun.

Schol-R-LEA wrote:Let me be even more blunt: QUALITY DOES NOT SELL SOFTWARE. Period. Marketing sells software. Microsoft, Oracle, IBM, Apple, Facebook, Google, and pretty much every other "successful" IT firm succeeded because they pounded people with the idea that they needed what they were selling.

Brendan wrote:Did you forget that we're talking about OS development projects; where it's almost always "single designer/developer creates everything"? For this context, almost everything you've said above is completely irrelevant.

Brendan wrote:Do you think that "same as everything you're already using; except less mature/stable/optimised, fewer drivers and no support" is an excellent way to convince people that they need what you're selling?

Brendan wrote:Have you considered that perhaps "does something that other OSs don't/can't support" is a much better way to convince people (that can benefit from whatever that feature is) that they need what you're selling because there is no alternative?

Schol-R-LEA wrote:

Brendan wrote:Did you forget that we're talking about OS development projects; where it's almost always "single designer/developer creates everything"? For this context, almost everything you've said above is completely irrelevant.

Only until it actually starts seeing use by others, and others begin contributing to the project. Or do you expect to write a fully functional, professional-quality system on your own? That hasn't been a reasonable proposition since the mid-1970s, and was only reasonable then because microcomputers were so limited in both hardware capacity and user expectations - in larger systems, one-man shows were a thing of the past by the time ITS was in the works, which was around 1966.

Schol-R-LEA wrote:

Brendan wrote:Do you think that "same as everything you're already using; except less mature/stable/optimised, fewer drivers and no support" is an excellent way to convince people that they need what you're selling?

I would normally say "of course not", except, that's exactly where Linux was until around 1998 or so, and where Linux desktop managers basically still are. Linux on the desktop may not be setting the world on fire, but it isn't something entirely negligible either.

Schol-R-LEA wrote:Don't misunderstand me: I want you to do something new, I am thrilled to see you and others here trying new things. My concern is that there is a risk in trying a new direction, and you seem to have an awful lot of your ego riding on the success of this one. The problem is that you are berating people whose interest is in learning, for following a well-trod path, which - correct me if I am wrong here - is exactly what people do when learning something new for which an established path exists. Most of the people here don't have the same goals you do, or I do, and arguing that a hobby project needs to break new ground just for the sake of innovation is pretty condescending.

Schol-R-LEA wrote:

Brendan wrote:Have you considered that perhaps "does something that other OSs don't/can't support" is a much better way to convince people (that can benefit from whatever that feature is) that they need what you're selling because there is no alternative?

I am not trying to sell anything, and to the best of my knowledge neither are any of the other members except you, rdos (who has been at this longer than any of us, I think, and has actually sold his system in the past), and perhaps embryo and onlyonemac. Most of the posters here are indulging in a mix of curiosity and tinkering, not trying to develop a product. While I agree that this is not a reason to slack off - one should try to fulfill one's goals to the best of their ability, and success can pick the least predictable times to strike (e.g., we're stuck with a lot of BS because Dennis and Ken didn't think their game system would ever be used outside of Bell Labs) - it doesn't mean that one's reach must exceeds one's grasp when working on something they are trying to do for fun.

Schol-R-LEA wrote:If your argument were directed towards me personally, I would see it very differently, because I do have ambitions for my work - ambitions of the sort that would lead to degrees and published papers rather than deliverable systems, but still - and, had you made this argument to me about my system, I would have deferred to your judgment (EDIT: sorry, that was a bit of a brainfart, I don't know why I originally used a word that meant the opposite of what I intended there). Thing is, this is a thread about a personal project meant to learn how something works. It isn't meant to see the light of day, and if it does it will be entirely by accident. It does not need to be held to the same standards you seek to hold yours to.

rdos wrote:The way the IT industry is working is the major reason why I wouldn't continue being in it unless I actually got paid for working on my OS. I'm certainly not learning anything about the new huge frameworks that big companies use to tie-up their customers. The wealth of hardware, standards, and secrecy about how hardware chips operate isn't encouraging. It's all part of tying people up so they have no time to develop better frameworks.

rdos wrote:It's big businesses that are setting the agenda, which include operating systems. You cannot create a successful operating system today unless you have the support of a huge number of hardware manufacturers that will write drivers for you. Bill Gates or Linus Torvalds would not have gotten their operating systems out today, and both of them were awful designs to begin with.

Brendan wrote: Yes; but in that case the goal becomes convincing hardware manufacturers to support the OS (rather than convincing end users). In the end it adds up to mostly the same thing - trying to do something that will convince someone.

rdos wrote:

Brendan wrote: Yes; but in that case the goal becomes convincing hardware manufacturers to support the OS (rather than convincing end users). In the end it adds up to mostly the same thing - trying to do something that will convince someone.

Yes, but that would turn me into a sales-person, not somebody designing and implementing an OS. I hate selling things, so that would make me uninterested in this endeavor. Besides, people that are good at selling their ideas often have bad designs and lousy products. Thus, I feel that if I'm successful at selling my design, the design would be lousy. Unless I have the backing of a large organization with sellers that are willing to listen to my ideas.

Brendan wrote:.......Yes; but in that case the goal becomes designing (and implementing an initial implementation of) an OS that's capable of selling itself. ......

tsdnz wrote:cloud

tsdnz wrote: because it is much faster ... it is for a cloud environment.

MichaelFarthing wrote:

tsdnz wrote: because it is much faster ... it is for a cloud environment.

Err...
Don't call us, we'll call you.

zenzizenzicube wrote:

tsdnz wrote:cloud

Do you mean that it's highly available and scalable, or do you mean that it's just to be thrown on servers somewhere?

set up UDP socket, etc..
while (true)
{
Send 4 bytes of udp
}

tsdnz wrote: 1. Kernel - This is just for ABIT, not available to users,

tsdnz wrote:It can handle 40 million small packets (64 Bytes) per second in plus 40 million out, total of 80 million.

Schol-R-LEA wrote:Sorry, ABIT?

Schol-R-LEA wrote:40,000,000 packets x 64 bytes x 8 bits = 20.48 Gbps, assuming that the 64 bytes includes the frames and no other traffic is occurring. Just how fat a pipe do you intend to feed to?

Schol-R-LEA wrote:Similarly, if you are sending packets at 12M per core, then assuming that your server is running at a consistent 4.5 GHz for every core, that gives you roughly 375 cycles per packet, assuming full load, no cache misses, and no other CPU operations on that core including scheduling and the I/O operations themselves. This puts you (to zeroth approximation and with some slop for pipelining) in the range of 300-400 instructions available for the processing of each packet, maximum. Care to reconsider that claim?

Schol-R-LEA wrote:The numbers aren't impossible ones, admittedly, but they seem rather improbable in the absence of more specific facts about the packet processing, packet contents and data sources, scheduling policy, I/O subsystems, and transmission speed of the networking hardware.