Social FLOSS “Out There”

Since RMS enunciated the principles of Free Software so many years ago, the world of IT has crossed time and space. “On January 14th, Twitter suspended @Barbijaputa’s account after she participated in a conversation about sexually transmitted diseases. The next day, she created a profile on GNU social node Quitter.se and started posting. Her Twitter followers proved willing to follow her all the way to GNU social, and began joining existing nodes en masse and starting their own.
The growth was so explosive that the some of the existing GNU social nodes were unable to handle the traffic.
See Thousands of Spaniards leave Twitter for GNU social — Free Software Foundation — working together for free software.”
While the original issues with users controlling software instead of being controlled by it still linger on the personal computer, servers, both local, remote and in the cloud introduce a new dimension.

It’s more efficient to have a few servers serving many thousands of people than having thousands of copies of some FLOSS. Indeed with networks of computers holding impossibly large databases, it’s not practical to have FLOSS on every PC taking care of business. One needs FLOSS on those servers to make them more trustworthy.

Social media is the extreme with hundreds of millions of users dealing with some central authority controlling every detail of a user’s experience. In Spain, the chickens have come home to roost and folks are giving FLOSS on federated servers a try. There will be teething troubles getting everything to scale but at least the users will have some say. Security is a whole other kettle of fish but security is a problem on the non-Free systems anyway.

I think this is an idea that can scale by eliminating bottlenecks. If there is no “central authority” there could be fewer bottlenecks. Can it be secured to prevent DOS and other baddies? Time will tell. The world can and does make its own FLOSS for PCs. Why not for networks of PCs and servers?

About Robert Pogson

I am a retired teacher in Canada. I taught in the subject areas where I have worked for almost forty years: maths, physics, chemistry and computers. I love hunting, fishing, picking berries and mushrooms, too.
This entry was posted in technology and tagged , , . Bookmark the permalink.

94 Responses to Social FLOSS “Out There”

  1. Deaf Spy wrote, “the world nowadays seems to praise and pay for intellectual labor much higher than material one.”

    I’ve earned money both ways in my career. Sports gods would disagree with you as would some dead painters…

  2. Deaf Spy says:

    Anyway, I’ve actually built stuff that met its physical requirements like that hoisting frame, a family swing set which lasted 20+ years and a host of other structures

    Mr. Pogson, you seem to completely reject the very concept of “paid intellectual product”. Your statement confirms that you count as paid labor only the one that results in material assets. This is very typical for Marx and communism.

    Strangely enough, the world nowadays seems to praise and pay for intellectual labor much higher than material one.

  3. oiaohm says:

    Robert Pogson the reality is people like DrLoser would most likely reinvent the wheel because they don’t go anywhere near Linux based solutions.
    http://cad.arcad.de/products_architecture_arcad.php

    Yes there are some very particular Closed source Linux Cads out there that are designed to cover particular types of construction way better than autocad.

    Autocad is attempting to use a jack hammer to put in a nail lots of the time.

    The annoying part is most of the industry targeted cads are OS X or Linux. With most being just Linux.

  4. DrLoser, like most losers has “uttermost contempt.”

    Anyway, I’ve actually built stuff that met its physical requirements like that hoisting frame, a family swing set which lasted 20+ years and a host of other structures. Even the birds like the houses I build for them. Meanwhile, the losers of the world rarely accomplish anything.

  5. DrLoser says:

    You may have noticed, Robert, that in the preceding discussion, I regard your understanding of structural engineering with nothing but the uttermost contempt.

    But just in case you failed to notice that small fact:

    I regard your understanding of structural engineering with nothing but the uttermost contempt.

  6. DrLoser says:

    Anyway, back to scaling “Quitter.”

    Not quite as hilarious as Dougie trying to understand how qualified engineers build a roof without taking their shoes and socks off.

    But almost.

  7. DrLoser says:

    Instead of using a computer, why not use an abacus?

    Who knows, Doug-nut?

    Are you proficient with either one?

  8. dougman says:

    Instead of using a computer, why not use an abacus?

    Loser is surely dimwitted and just mad that he lost his pension and that David Cameron is spending his money.

    Doesn’t Loser remind you of Piers Morgan?….FACTS just bounce right off him, as he has an agenda to push.

    LINUX IS TO COMPUTING WHAT FREEDOM IS TO MANKIND!

  9. DrLoser says:

    It takes as long to enter data into software as it does to do the lookups so software is not all that wonderful.

    No, you’re correct, Robert.

    Why bother with FLOSS software, which is not all that wonderful, when a slide rule would do just as well?

  10. DrLoser says:

    Anyway, Robert, that Fred Brookes brain-fart.

    Are you going to pretend you never said it, or are you going to apologise, or are you going to provide a justification?

    Just out of interest.

  11. DrLoser says:

    GMSH can do just about anything.

    A 3-D mesh program that can do “just about anything,” Robert? It’s a tricky field, is 3-D meshes. Perhaps somebody more competent than you or I can judge on its 3-D meshiness.

    As regards engineering? No, it’s hopeless.

    Sad to say, my company even found a better mesh library than this one. It’s not really going anywhere as far as I can see.

  12. DrLoser says:

    but, I guess that doesn’t matter to DrLoser. He’s drunk the KoolAid…

    Or alternatively work in a field where none of these miserable little efforts ever surface in conversation, Robert.

    Let’s just start with Load Transfer Sites.

    Show me a single one of those that does such a thing.

  13. DrLoser says:

    But we digress.

    Now, Robert. Engineering aside. How do you propose to attract the roughly 1+ billion daily users of social marketroid sites such as Facebook or Twitter to FLOSS?

    Please tell me that it’s not some overwhelmed, doomed, feeble attempt from the FSF like Eating your own toe fungus, on screen! A permissive legal license.

    We can discuss the idea of “federated servers” later.

  14. DrLoser wrote, “There are no FLOSS competitors in the domain of roof design. I don’t just mean “none that a roof designer would use.” I mean none.”

    Uh, Continuous Beam Analysis, Elmer finite element physics and GMSH can do just about anything.

    but, I guess that doesn’t matter to DrLoser. He’s drunk the KoolAid…

    Chuckle. Here’s a guy designing trusses with OO/LO Calc

  15. DrLoser says:

    Any interesting examples of FLOSS structural engineering programming suites to offer?

    No, I thought not.

    But, here, let me help you out.

  16. DrLoser says:

    Loser, seems to think that all this engineering needs to be done over and over and over again. When in fact, any competent person could spit numbers out all day sans any software.

    Out of the mouths of babes, sucklings, and snake-oil salesmen!

    Every single bit of that is unquestionably correct, Dougie.

    That over and over again bit? It may surprise you to know that over and over again is what computers do.

    Thus why people buy commercial software to do it. Not everybody’s time clocks in at $0 per hour like yours, Dougie. Even assuming you could do the engineering calculations, which, believe me, you can’t.

  17. DrLoser says:

    e.g. Suppose the outfit did the CAD in the old days and has a library of designs for roof trusses that do 98% of the jobs.

    I will therefore so suppose.

    A roof truss is not much more than a bunch of wooden bits planked together, Robert. Or nailed. Or webbed. Or plated. Or bolted. Or screwed. (Roughly in order of industry preference.)

    Any old fool can design a roof truss in isolation, Robert. Such old fools would be massively assisted in their defective designs by the unfortunate fact that, well, carpenters ain’t so smart, and that therefore building codes worldwide have something approaching a 400% tolerance for lousy design on a truss.

    You don’t need CAD-CAM to build a truss.

    Unfortunately for your argument, whatever it is (FLOSS is pointless? Slide Rules are Good? Who knows, at this point?), one single truss does not a roof make.

    And then we go into load transfer points, static models, exposure models ….

  18. dougman says:

    Loser, seems to think that all this engineering needs to be done over and over and over again. When in fact, any competent person could spit numbers out all day sans any software.

    BTW: the building design I am putting in was engineered years ago, so why should the manufacturer recalculate everything? My slab is being over-engineered by me, but is being stamped by a CE I know for $100 and will be approved for permitting.

    No need for $$$$$$ software to be honest.

  19. DrLoser says:

    It takes as long to enter data into software as it does to do the lookups so software is not all that wonderful.

    And the whole point of this site would be?

  20. DrLoser says:

    Oh well, I should be as fair as it is humanly possible to be fair to oiaohm.

    This is a Friday, a traditional day for such things. oiaohm has gone out of his way to be relatively sensible. Robert appears to have fallen off a cliff in that respect. Therefore:

    Mythical man-month?

    BWAHAHAHAHAHAHA!

  21. DrLoser says:

    I also encourage you to admit that bringing Fred Brookes into the conversation was a momentary “whoopsie, didn’t really think that one through.”

    Of course, you might have thought it through. Perhaps you have a good reason to suppose that Fred Brookes is appropriate in this context?

  22. DrLoser says:

    You presuppose productivity does not decline. It may well decline with one less body replaced by one CAD package.

    You’re just arguing for the sake of it, aren’t you? I’m not arguing about productivity “declining.” I’m arguing about slide-rule productivity being replaced, with some considerable ease, I might add, by CAD-CAM productivity. And it’s not a difficult argument to make.

    God knows, it’s not like the world has an overabundance of engineers. I’m pretty sure those guys and gals could be gainfully employed elsewhere.

    Now, as to “productivity,” Robert. Are you seriously proposing that FLOSS structural engineering programs are barely above the level of the slide-rule?

    I mean, I think they are. But it’s a very strange argument for you to make, isn’t it?

  23. DrLoser says:

    Joe has a house 31 feet wide and Bob has a house 32 feet wide. The same trusses may or may not work for both. Then Bob may live where snow is expected at the rate of 4 ft/annum while Joe gets 1 ft. Bob may live in a place with several instances of 80km/h wind while the other guy rarely exceeds 50km/h.

    Somehow you have perceived the truth of what my employers have not, Robert. Towering Genius!

    Or possibly not.

    Let us try again. Without a commercial package, Joe is assigned Engineer A and Bob is assigned Engineer B. Engineers A and B take these wildly different requirements, and, all on their tods, come up with the two solutions.

    Or alternatively we dispense with the services of Engineer B and replace him with the commercial software package. Here’s what we do:

    1. Roof and truss design, via click-and-drag (standard AutoCAD stuff): 30 minutes, tops, per roof.
    2. Building codes and all the engineering numbers and calcs: built in.
    3. Snow load: as I say, a drop-down box for each. You failed to mention the duration of snow load, which is another drop down box. (For loads in general, you get to choose between momentary, short, medium, long and permanent. It depends on the load. That too is factored in. I’m not going to confuse you with the idea of “load combinations,” but trust me this is another thing that commercial software sorts out for you.)
    Let’s be fair. Thirty minutes for an arthritic old man to finally select the correct snow load.
    4. Wind load. See snow load.

    Hmm, let’s see. That’s about two hours’ worth of a roof designer on a computer using proper software, not floss-dross.

    Ample time to replace the other engineer with a kid who likes to design pretty decorations on the ends of trusses using a pricey Mac-Air, I think.

    Sorry, Robert — your numbers don’t add up. In many cases, they will. In this case, never.

  24. DrLoser says:

    I haven’t even scratched the surface here, Robert.

    Load transfer sites?
    Cycling plates and/or timber for minimum cost/maximum safety?
    Nailing patterns? (Yes, nailing patterns are important.)
    Load deflection?
    Vibration?
    Service level stress versus Ultimate level stress?
    Choice of materials? (wood, glulam, etc)
    Perpendicular tension? (Don’t want them narsty nails to rip through the grain.)

    I’m a neophyte in this, Robert. I could probably name at least five more things that apply to roof design in a purely engineering sense.

    And then you’ve got the design details. And the on-site details. And the costing. And the inventory.

    No, sorry, Robert. We are no longer Prairie Pioneers.

    Most of us want our roofs built, cheaply and reliably, by professionals who routinely use commercial software to do so.

    Slide-rules don’t cut it in the modern world. And I don’t really see how FLOSS ever will, either.

  25. DrLoser wrote, ” The data is inherent in the materials (and of course the building codes).”

    Nonsense. Joe has a house 31 feet wide and Bob has a house 32 feet wide. The same trusses may or may not work for both. Then Bob may live where snow is expected at the rate of 4 ft/annum while Joe gets 1 ft. Bob may live in a place with several instances of 80km/h wind while the other guy rarely exceeds 50km/h. One house may face South or be near the ocean or… An engineer will take into account all of it and no application is going to have a “one click” solution for every case.

  26. DrLoser wrote a lot in bold text. He also wrote, “If one of them is a superstar, and is prepared to work at the same pay, then fine, no problem. That one will do double the work, for some inscrutable reason.”

    It’s not about how the un-CAD guys work but about the productivity with the CAD system. You presuppose productivity does not decline. It may well decline with one less body replaced by one CAD package. You presuppose productivity will double per human with CAD. That could happen, or not.

    e.g. Suppose the outfit did the CAD in the old days and has a library of designs for roof trusses that do 98% of the jobs. Replacing one body with CAD probably won’t change a thing. Suppose the outfit does 90% custom work of a kind never seen before… CAD may well multiply the performance of a user 100-fold but they may well have to hire an engineer to check all the results, a great increase in the payroll.

  27. DrLoser says:

    I’ve seen several houses built and never felt the need to buy specialized software. In at least one case, the architect did work with pencil and paper and the engineer did certify the design. I think the total cost to us for both was ~$1500 and the process was done in a couple of weeks.

    Well, let’s just cost that one out, Robert. And let’s ignore the fact that this is anecdotal evidence. And let’s assume that you saw the Best of the Best of the Possible Best for that $1500.

    So, we’re talking two weeks of an engineer’s time. And two weeks of an architect’s time.

    That’s four man-weeks’ time for a gross of $1500. Or $375 each per week.

    I implore you, Robert, send me the e-mail addresses of these poor souls.

    I can get them a job as a fry cook that will pay better.

  28. DrLoser says:

    This is the “mythical man-month” stuff again. Not all designers are created equal…

    What a preposterous lemma.

    First of all, I am presupposing two identically qualified roof designers. I don’t care what standard they are at.

    If both of them use slide-rules, you need to pay both of them at a notional $30,000 per year. Which cost I have intentionally understated.

    If one of them is a superstar, and is prepared to work at the same pay, then fine, no problem. That one will do double the work, for some inscrutable reason.

    But if both of them are equally able, then it’s cheaper to replace one of them with commercial software. This is not at all hard to understand.

    Now, let me ask you three questions on your ridiculous side-track here, Robert.

    1. Have you read Fred Brookes’ book?
    2. If so, do you understand the fundamental cost failure that he identifies?
    3. If so, how can this possibly apply to removing a programmer/engineer/designer/whatever and replacing them with a valid commercial software package?

    I mean, you’ve made some preposterous claims in the past, Robert. But this one is not merely preposterous — it verges on evidence that you haven’t really bothered to think things through at all, doesn’t it?

  29. DrLoser says:

    It takes as long to enter data into software as it does to do the lookups so software is not all that wonderful.

    Foolish speculation, if you ask me. To start with, if you use a commercial CAD-CAM system to design a roof, guess how much data you need to enter?

    Go on.

    None, that’s what. The data is inherent in the materials (and of course the building codes).

    Let me patiently try to explain this to you, Robert. I, who am not a roof designer and am most certainly not a structural engineer, could design a roof using our software or any other commercial alternative without once having to resort, either to data entry, or to a calculator. And that roof design would pass the on-site inspector in a heartbeat.

    I tell a small fib about the data entry. There are drop-downs for timber grade. There are drop-downs for metalwork. There are tick-boxes for, say, inside or outside exposure models. On important things like snow load or wind load, you can enter zone information (which building inspectors will respect) or even an arbitrary value in newtons per square meter (which building inspectors will obviously check), and get the rest done for you. Believe me, the mucking around that the package has to handle just to get French wind loads to pass inspection is something so horrible that you would never want to do this via a slide rule.

    I’m going to leave out the details of the static model.

    Now, once you’ve got the design and engineering done, you still have to produce the various cutting outputs, customer outputs, production outputs, site outputs and so on. And you’d better believe that any single one of them needs to be able to be exported into one of at least five formats. Because they do.

    And then there’s the inventory.

    And then there’s the costing and the boring accountancy bits.

    I can just about imagine an obsessive-compulsive lunatic engineer who has lost his marbles wanting to do the engineering bits on a calculator, Robert — or the moral equivalent, which would be some horrid lash-up of buggy FLOSS crap — but I can’t really see an engineer want to do all the rest of that on their own.

    Can you?

  30. DrLoser says:

    Engineers love doing calculations by table lookup rather than first principles.

    Then you know some seriously deranged engineers. Any engineers who have a degree (a simple Batchelors will do) acquired since about 1990 love using software that does the bulk of that for them. A good 90% at least of the calculations involved in roof engineering is sheer pointless tedium.

    Now, do you mean that engineers love pulling out the calculator (or an excel spread-sheet) and challenging the one or two elements in the remaining 10% that don’t look right?

    Yes, Robert, engineers love to do that. And you know why? Because they are not the pointless drones you seem to wish they were. I have seen roof engineers do this several times with our software, because that is what they are paid to do. Like building inspectors, they are paid to use their expertise to question things that do not, on the surface, add up.

    It’s akin to your weirdo notion that handing in bug reports for FLOSS software is actually a valuable thing to do, except that, and in contradiction to such a complete waste of everybody’s time, it is actually valuable.

    I’ve had to correct one bug in metalwork over the last year or so, based on a correction by a professional with an excel spread-sheet.

    I’ve had QA throw about ten tickets back at me for correction, because our QA is actually qualified structural engineers with calculators.

    This is actually part of the process.

    Working engineers’ fingers to the bone on pointless calculations, or asking them to accept sub-standard crap like FLOSS, is not part of the process.

  31. dougman says:

    re: I’ve seen several houses built and never felt the need to buy specialized software. In at least one case, the architect did work with pencil and paper and the engineer did certify the design. I think the total cost to us for both was ~$1500 and the process was done in a couple of weeks.

    I have seen that as well here. My building is utilizing prebuilt trusses suitable for locale snow/wind loads and utilizing hangers. Actually, I am spending more time on the column footers, integral concrete slab footers, basalt rebar spacing, slab insulation, drainage and control joints to alleviate any frost heave.

  32. DrLoser wrote, “either way, would be doing a lot of calculations by hand.”

    Engineers love doing calculations by table lookup rather than first principles. It saves a lot of time. That’s what the software is replacing. It takes as long to enter data into software as it does to do the lookups so software is not all that wonderful. Of course, an engineer can set up defaults and stick to them to save all that but if he’s going to check a drawing/evaluate a plan, he may not be using his defaults. e.g. I’ve had an architect create blueprints of our home and an engineer had to check them and certify them acceptable. That requires working with whatever is on the plan, not the usual defaults of the engineer.

    DrLoser also wrote, “If you employ two roof designers, working either by hand or via non-performant software, you’re spending a bare minimum of $30K per year for them. If you can replace one of those roof designers with software that doubles productivity at a cost of even $15K per year (my company supplies its software free, but whatever), then you win.”

    This is the “mythical man-month” stuff again. Not all designers are created equal… Then there’s lock-in and forced upgrades. The cost of software is not only the capital cost. I expect a lot of design software will be used rather infrequently creating a library of designs which are pulled off the shelf as needed. There will be times when a custom design is needed but the expert designer will know how to scale things too. It might be a better measure to use $/order rather than the capital cost of the software or salary of the worker. A high-volume shop will be able to distribute whatever costs over many units. A small shop might do the maths DrLoser’s way, constrained by a limited budget. I’ve seen several houses built and never felt the need to buy specialized software. In at least one case, the architect did work with pencil and paper and the engineer did certify the design. I think the total cost to us for both was ~$1500 and the process was done in a couple of weeks. The only things I didn’t like about the resulting house were that it was not square/optimal (for cost/heating/maintenance) and there were two or three lapses in workmanship of the builders. The Little Woman overruled me on the first and bad things do happen in the real world. It was a wonderful home in which we raised three great kids.

  33. DrLoser says:

    Using proper software, you’re also far more likely to get your roof signed off by an inspection engineer than if you waved around a bunch of schematics that came out of the back end of a FLOSS Pantomime Horse.

    As always with the commercial world, Robert, time is money.

  34. DrLoser says:

    …such as Maximum Span Calculator for Wood Joists & Rafters. There are even versions for Android/Linux…

    Whoop-de-doo, Linux can manage a simple look-up for maximum span, based on grade and depth of timber. That’s some awe-inspiring stuff right there.

    Engineering even the simplest of wooden roofs is a little more complex than that, Robert.

    All designs according to building codes in Canada have to be certified by a proper engineer but there’s no rule he/she can’t use such software. In the old days, they used slide rules or pencils.

    And, either way, would be doing a lot of calculations by hand.

    I’ve said it before, Robert. If you employ two roof designers, working either by hand or via non-performant software, you’re spending a bare minimum of $30K per year for them. If you can replace one of those roof designers with software that doubles productivity at a cost of even $15K per year (my company supplies its software free, but whatever), then you win.

    There are no FLOSS competitors in the domain of roof design. I don’t just mean “none that a roof designer would use.” I mean none.

    It’s just not one of those areas where FLOSS is ever going to make an impact.

  35. DrLoser, implying no engineering can be done with a ChromeBook wrote, “First of all, there’s presumably no engineering at all in that building, which means that as a professional roof designer (etc) you’ll be wasting wood at the least.”

    That’s just silly. GNU/Linux can run all kinds of engineering software and so can many web applications such as Maximum Span Calculator for Wood Joists & Rafters. There are even versions for Android/Linux… All designs according to building codes in Canada have to be certified by a proper engineer but there’s no rule he/she can’t use such software. In the old days, they used slide rules or pencils.

    Heck, one can build roof trusses by formula/table lookup with the engineering built in.

    See, for example, Roof Truss Design with Nailing Schedules.

  36. DrLoser says:

    Eh, I made up some nice CAD drawings for a building I plan on building. Did so on a Chromebook with LINUX!

    First of all, there’s presumably no engineering at all in that building, which means that as a professional roof designer (etc) you’ll be wasting wood at the least.

    And secondly, if it falls down, it only falls down on top of you. Which is still regrettable, but you’re not going to sue yourself, are you?

    Not gonna hack it in the outside world, I’m afraid.

  37. Adam Queen says:

    oiaohm can I have the filter too? Thanks.

  38. dougman says:

    Eh, I made up some nice CAD drawings for a building I plan on building. Did so on a Chromebook with LINUX!

    I also use FreeCAD on my Linux Mint desktop.

  39. DrLoser says:

    Oh, and by the way.
    No, some amateur lash-up social network in Spain cannot “transcend the bounds of time and space.”

    Nice try, though.

  40. DrLoser says:

    In the real world, hot damn, yes, we need “every stress accounted for.” We don’t just “make do” with a welding gun after the event.

    Robert Pogson apparently believes that professionally qualified engineers, world-wide, employed by companies who live and die on a profit-and-loss basis, have no need whatsoever to make sure that “every stress is accounted for” and should resort to a rugged plains frontiersman armed only with a welding gun and a wad of duct tape.

    Robert Pogson would be laughed offsite by even the simple souls who stitch wooden trusses together to form a bog-standard roof.

    This isn’t the stone ages, Robert.

    People use commercial CAD applications for even small engineering problems these days. Why? Because it pays to do so.

  41. DrLoser says:

    Robert Pogson wrote, in re surge protectors, that I “apparently [failed] to read that the boss forbade such sensible precautions.”

    Robert Pogson apparently failed to read the part where I pointed out, very clearly indeed, that this is not something you can blame upon the months and months of work spent on the CAD system in question.

    Let alone use as an excuse not to use a professional CAD system in 2015, thirty five years later.

  42. DrLoser, going off on another tangent, wrote, “it’s a damn good reason why no reputable company in the Western World is going to touch sub-standard, broken, unsupported, uninsured, FLOSS crap.”

    Yeah, like NYSE, Peugeot, IBM, etc. [SARCASM]

  43. DrLoser wrote, “I suppose the plausible solution of a surge protector never occurred to any of the geniuses and their months-long toil on a CAD solution, did it, Robert?”, apparently failing to read that the boss forbade such sensible precautions.

  44. DrLoser says:

    The problem was that in the real world sparks happen and we regularly wrecked expensive HP power supplies.

    I suppose the plausible solution of a surge protector never occurred to any of the geniuses and their months-long toil on a CAD solution, did it, Robert?

  45. DrLoser says:

    Now, if you were wanting to build a skyscraper and have every stress and cost accounted before starting ordering and fabricating…

    I’ve just noticed, Robert. Let’s generalise from “a skyscraper” to any sort of engineered object whatsoever.

    Your proposition (“let’s not bother about it until we need do”) is diametrically opposite to the real-world, engineering way of doing things.

    In the real world, hot damn, yes, we need “every stress accounted for.” We don’t just “make do” with a welding gun after the event.

    And even when you’re talking about on-site adjustments (a site-cut, an inconvenient chimney, whatever), these are also pre-factored.

    And, much though you loathe and despise accountants: yes, on the whole, they do tend to prefer knowing the cost of an engineering construct ahead of time.

    This is why corporations pay good money for proven, working, tested, commercial CAD-CAM solutions, backed with insurance and other legal guarantees, Robert.

    And it’s a damn good reason why no reputable company in the Western World is going to touch sub-standard, broken, unsupported, uninsured, FLOSS crap.

  46. DrLoser says:

    German U-boat Captain: I am making notes, Captain, and your name will go on the list; and when we win the war you will be brought to account.
    Captain Mainwaring: You can write what you like, You’re not going to win the war!
    U-boat Captain: Oh yes we are.
    Mainwaring: Oh no you’re not.
    U-boat Captain: Oh yes we are!
    Pvt. Pike: [Singing] Whistle while you work, Hitler is a twerp, he’s half-barmy, so’s his army, whistle while you work!
    U-boat Captain: Your name will also go on the list! What is it?
    Mainwaring: Don’t tell him Pike!
    U-boat Captain: Fifi! My Darling! Love of my life!

    … und so weiter.

  47. DrLoser says:

    Why you used the spam attack method to attempt to get me to say what you wanted oldfart.

    Is that supposed to mean anything, Peter Dolding?

    Anyway, back to SCSI and ESXi. Your expert google-splatting on that?

  48. oiaohm says:

    oldfart I will not do you that option. Mind you its not impossible to get me to remove a person from my blacklist.

    Why you used the spam attack method to attempt to get me to say what you wanted oldfart.

  49. oldfart says:

    “Really there are only 4 here from the old TMR.”

    Do me a favor and add me as well. Then we can get on with our talkback with Robert Pogson without having to wade through your irrelevant crap.

  50. oldfart says:

    “And, obviously, everybody on earth was within easy range of the computing power of, say, a Pentium V. (Note to subs — check that one.)”

    By the late 80’s You had access to 386DX class systems in the 10’s of megahertz, or M680xo Class with mhz clocks. Both 32 bit capable systems. The 486 class systems came about by the early 90’s and pentium class from the mid 90’s on.

  51. oiaohm says:

    LinuxGentlesir mind you now that I have a filter that works maybe I should publish it. I currently I have two people in the filter kurkosdr and DrLoser.

    Really there are only 4 here from the old TMR.

  52. DrLoser says:

    I think the right thing is Dr avoiding you btw

    Wise advice, Kurks. I’ll try to concentrate on the other loonies.

  53. DrLoser says:

    Now, this “Social FLOSS” of yours, Robert.

    How do you see it scaling beyond the level of, as Kurks points out, a small Spanish village?

  54. kurkosdr says:

    I just noticed I forget to mention here that I had blacklisted DrLoser. Note I have done it by greasemonkey so none of his posts exists to me any more.

    This is not TMR kid, where most users have grasemonkey filtering out the same spamlist. Here, everyone else is going to see DrLoser’s posts, reply to them and quote them, so your plan to avoid DrLoser* will fail, you will just have some gaps when it comes to the whole discussion stream.

    Not that it will affect the quality of your posts in the slightest bit, however. Guess why.

    *I think the right thing is Dr avoiding you btw

  55. DrLoser says:

    Let’s just theorise here.

    Let’s take Mo and Joe, two structural engineers (it happens to be my current speciality. Substitute any other engineering discipline you prefer) who are paid a miserable pittance for their expertise. Let’s stipulate $30,000 each.

    Let’s go Hog Wild and spend $600 on a professional CAD-CAM system.

    Sadly, we are going to have to let Mo go. Nothing to do with anti-Semitism.

    It’s just that $600 is demonstrably cheaper than $30,000

    Next stupid objection, please.

  56. DrLoser says:

    Oh well, back to somebody with relevant experience from the 1980s. When men were real men. Small furry creatures from Alpha Centauri were real small furry creatures from Alpha Centauri.

    And, obviously, everybody on earth was within easy range of the computing power of, say, a Pentium V. (Note to subs — check that one.)

    After months of CAD we had a design that our whole “machine development” group thought would work, was feasible and was affordable. However, the boss insisted that high voltage power supplies involved should have no surge protection so that the voltages in the high vacuum would be “correct”. The problem was that in the real world sparks happen and we regularly wrecked expensive HP power supplies.

    And the problem would have been resolved by ditching CAD altogether, and relying on the time-honoured abacus method, Robert?

    Spiffy!

  57. DrLoser says:

    No, Fifi. You have googled and don’t have a clue. I have worked.

    Apparently you’ve wandered onto the wrong site here, then, mate.

  58. DrLoser says:

    I think it’s pretty obvious based on his incredibly frenzied, conceited and timecube-like posting style that he has some undiagnosed mental disease. Something more insidious then harmless eccentricity.

    Out of the faintest possible interest, Adam, which quotes from Lessig do you particularly recommend for Robert’s delectation and delight?

  59. DrLoser says:

    I just noticed I forget to mention here that I had blacklisted DrLoser. Note I have done it by greasemonkey so none of his posts exists to me any more.

    Well,, it’s no fun arguing with the Ravenous Bug-Blatter Beast of Trial, Fifi.

    Truth is Beauty, Beauty is Truth.

    And then the rotten thing eats you for breakfast. It’s a tough little world, selling your mediocre physical wares under a lamp-post in the outback, isn’t it?

  60. LinuxGentlesir says:

    oiaohm,

    I think it’s pretty obvious based on his incredibly frenzied, conceited and timecube-like posting style that he has some undiagnosed mental disease. Something more insidious then harmless eccentricity.

  61. oiaohm says:

    I just noticed I forget to mention here that I had blacklisted DrLoser. Note I have done it by greasemonkey so none of his posts exists to me any more.

  62. DrLoser says:

    Now, if you were wanting to build a skyscraper and have every stress and cost accounted before starting ordering and fabricating, certainly AutoCAD or other packages would be extremely useful but for a huge number of projects/processes much simpler software is sufficient.

    Practically the only thing my company does is to help people design wooden trusses for roofs, Robert. Hardly at the same scale as designing sky-scrapers, is it?

    And yet, strangely enough, we still need CAD. It’s specialised to our purposes, and the engineering side of it doesn’t involve sticking a probe into an aquarium full of electrolyte — just a simple matter of thousands of result points in a static model, that’s all, could easily be done by a team of drafts-peeps using vellum and goose quills — but somehow or other we still can’t get away from that dratted RealDWG.

  63. DrLoser says:

    Always nice to see enterprising governments like Vorarlberg and Solothurn being cited as compelling evidence that “in fact there are more active deployments of both Grass and Qgis than Arcgis.”

    Me, I can’t get enough of Vorarlberg and Solothurn. Today, the Alps — Tomorrow, the World!

  64. DrLoser says:

    Deaf Spy if you type my name as FIFI again I will type you handle from then on as FIFI.

    Surely that would be quite confusing for the rest of us, Fifi. I mean, how would the rest of us tell you two apart?

    Ah, wait, I know. One writes fluently, talks sense, and has actual IT experience in the areas he talks about.

    The other one is a google-splatting functional illiterate. By the way, your cite was underwhelming in the extreme:

    Satisfactory performance: for most needs the lean MapServer or GeoServer perform better than full equipped ArcGIS Server.

    That, and a smaller disk footprint for a given dataset, will get you a cup of Joe, although not the froth on top. No matter. Froth isn’t a problem for you, is it, Fifi?

  65. oiaohm says:

    Deaf Spy if you type my name as FIFI again I will type you handle from then on as FIFI.

  66. oiaohm says:

    No, Fifi. You have googled and don’t have a clue. I have worked.
    Deaf Spy if you had a clue you would have never mentioned mapwindow. Why QGIS is the dominate.
    http://en.wikipedia.org/wiki/MapWindow_GIS
    Mapwindows is .net crap the USA Government decided to back.
    http://www.ga.gov.au/data-pubs/web-services/using-wms-and-wfs-in-qgis

    Accessing Australian data I have two options qgis or ArcGIS and if their is a issue you will be told to use Qgis. EU the same China the same Russia the same.

    Mind you all the USA servers work with Qgis and ArcGIS.

    Basically deafspy you have no clue. What the USA government does is not always in alignment with what everyone else is doing.

    QGIS is the globally support. ArcGIS you can have connection problems unless you have spent top dollar.

    If you are claiming worked DeafSpy you have only worked with in a minor percentage of the world data on the subject.

  67. Deaf Spy says:

    deafspy you have googled and don’t have a clue.
    No, Fifi. You have googled and don’t have a clue. I have worked.

  68. oiaohm says:

    LOL Mapwindow that is a Windows only product that is true crap.
    http://www.qgis.org/en/site/ this is the cross platform one that works.
    http://alternativeto.net/software/arcgisdesktop/
    Come on really how hard it is to use something like this deafspy.

    GRASS GIS and QGIS both have governments using them. In fact there are more active deployments of both Grass and Qgis than Arcgis.

    http://www.digital-geography.com/arcgis-server-vs-open-source-gis-solutions/

    Grass and Qgis support the open source servers.

    I am not into CAD, but I have quite some idea about GIS.
    deafspy you have googled and don’t have a clue. 2014 was the major change in deployments of GIS solutions. In fact back as far as 2009 was open source GIS being deployed ahead of closed source GIS.

    Mind you udig is java based option for GIS as well.

    GIS is one area where you have 2 functional open source server options. 3 functional desktop interface options. Worse 2 of the open source desktop interfaces for functionality review better than all the commercial competitors. Of course in this case using the open source solutions does not equal having to use Linux.

  69. Deaf Spy says:

    I am not into CAD, but I have quite some idea about GIS. There, ESRI ArcGIS is what AutoCAD is to engineers. The One and Only suit. There are some FLOSS alternatives (MapWindow), which even I can tell you suck. I have even fought with the source of MapWindow and DotSpatial to fix some bugs, and I can tell you, the code quality is generally low. Overengineered, with too many OO polymorphism here and too little there, with a number of half-assed implementations and buggy morasses that end up with out-of-memory errors. Good for a test and to do some simple plugin, but not much more.

    Btw, ESRI’s original file format, Shapefile, is pretty well documented and open. Their newer file formats – ESRI TIN, ESRI Raster and ESRI FileGdb – not so much. Still, despite Shapefile being open, no one managed to develop a sensible competitor to ESRI.

    Why? Because the actual file format doesn’t mean anything at all. What counts is the functionality (bloat!) and vision for development of this functionality. ESRI has both. The devs of MapWindow and DotSpatial still can’t handle large feature data properly, let alone develop something really useful for the engineers out there.

  70. kurkosdr wrote, “I am very curious to find out what kind of project is that which needs CAD, but is not using AutoCAD.”

    I’ve done CAD for many years and I don’t ever recall having even seen AutoCAD. It wasn’t even released until 1982, about the time I went to work in Saudi Arabia. In those days, “design” was done on paper and drawings required a draftsman who worked on large sheets with paper and pencil or pen for publication. The C in CAD in those days was for “Computer” but there wasn’t a particular need to combine drawing and that computerization since we stuck to well-defined mathematical curves more or less. They were fairly easily generated by lines and arcs on paper. In particular we worked with designs of electric and magnetic fields and needed shapes perpendicular to the fields we required. Finite-element analysis of one kind or another was often used but not essential. Often an analytic solution was available and the drawing just mirrored that. More complex systems were designed just with more pages of drawings rather than more complex drawings. A few people with the right skills could pull it all together. The modern applications embed those skills into the software permitting more rapid turnaround by fewer people but in the specialized fields requiring CAD it’s not likely any software can be guaranteed to replace the skill sets of even a small number of people. The best that can be said of them is that many steps are automated.

    An example: Suppose one needs a particular electric field shape. There’s only so much you can do because of the laws of physics. Symmetry alone may be enough to predict the shape of the field. The automation may only make more certain the magnitudes and those can be set digitally or by turning a knob somewhere. e.g. We sometimes used an aquarium containing some electrolyte. Potential shapes of metal could be formed and inserted in the tank. A probe could measure the electric potential field between shapes electrified. If the field distribution was incorrect, a second shape could be tried and by interpolation the correct shape could be determined in short order as all solutions would be a linear combination of a small number of potential solutions. When computers became sufficiently flexible, we replaced these analogue method with digital solutions of differential equations. We automated the maths rather than the design. One could quickly set some boundary conditions in the array and when the solution was correct, any equipotential surface was a valid solution of what shape of electrodes to use. Symmetry often allowed 1/4 or 1/2 the solution to be sufficient to solve the whole problem.

    Similar methods apply to strength of material design problems or optimization of the cost of production or even aesthetics. In my work we often produced 3-D objects to produce the desired effect in 2 or 3 dimensions. A few iterations of the design cycle usually converged on something easy enough to machine/afford/assemble. Embedding this intelligence into software may well speed up the process but the same kind of intelligence is still required by the human managers/engineers using the system to evaluate/choose designs so unless the project is very complex there may be no advantage at all. Given that many different solutions have similar steps/characteristics, software is certainly feasible but the creative ability of humans is still needed and humans only work so fast.

    Another example. After months of CAD we had a design that our whole “machine development” group thought would work, was feasible and was affordable. However, the boss insisted that high voltage powersupplies involved should have no surge protection so that the voltages in the high vacuum would be “correct”. The problem was that in the real world sparks happen and we regularly wrecked expensive HP power supplies. The record was 7 in one day… No matter how fancy the CAD, humans can still mess things up or improve on them. We had several members of the group aware of the flaw, including me but we were out-ranked and the misuse of PSUs continued for months until the techs repairing them objected… and reached higher powers by a more direct route than the chain of command. In the end everything worked beautifully but any speed up by automating CAD would have been of no benefit. Assuming fully automated CAD, the major expenditure of time and money for the project was acquiring material and machining it. CAD done in 1ns would not have brought the project to fruition even a week sooner. The electrodes were all carefully machined from OFHP copper and from the time the design was complete until fabrication was many months.

    Now, if you were wanting to build a skyscraper and have every stress and cost accounted before starting ordering and fabricating, certainly AutoCAD or other packages would be extremely useful but for a huge number of projects/processes much simpler software is sufficient. All we used was a couple of compiler/assemblers, mini-computers and a mainframe and some graphing software. Essentially we wrote the solution ad-hoc in a few hours of even less and the major time of computation was determined by the speed of the hardware, not the sophistication of the software. We had dozens of physicists and engineers who had the knowledge and skill to do the calculation so no expert software was required. We rarely needed more than 1K lines of code to do such work, and often just a hundred or so. I think the biggest code I ever used in this work which was state of the art at the time was 4K lines generated elsewhere. I rewrote it in 1K. That one did take a week or so to do but there wasn’t much creative about it just standard solutions to standard problems. The major work of physicists is to look at the most complex problems as a finite list of smaller/simpler problems. Engineers work in a similar fashion choosing not to reinvent the wheel.

  71. oiaohm says:

    Just so everyone reading this post knows due to DrLoser constant determination to use FIFI instead of Oiaohm to refer to me there will be no answers coming to the items DrLoser has raised if you wish for me to answer them please do your own posts. DrLoser is on my ignore list.

  72. DrLoser says:

    There are a lot of professionals on the planet and a lot of them do CAD but they don’t use AutoCAD. Same applies for MATLAB. Thank Goodness real professionals have choice.

    Thank goodness real professionals have choice indeed, Robert. But not a choice that you would give them, is it? You would label them Drug-Dependent Maniacs and deprive them of that choice.

    For what it’s worth, very few “real professionals” would waste their time on a FLOSS alternative to AutoCAD. I know this, because I work in that sector.

    As for MATLAB? Not in my field. But there seems to be an awfully large number of job advertisements out there that require MATLAB expertise.

    Not so many for Octave, Sage or Scilab, for some inscrutable reason.

    Possibly because all three are crap?

  73. DrLoser says:

    dwg at the start nothing about the format at all was documented. .cwp is in the same class.

    So what, Fifi? The fact remains that commercial organisations, even Freetarded commercial organisations like mine, use RealDWG and not LibreDWG.

    One assumes there is a good reason for these people to ignore their basic prejudice against paying for anything. What could it be?

    Just possibly, LibreDWG is crap.

  74. oiaohm says:

    .doc and .ppt the OLE format of them was always open. PSD has always been documented by adobe just exactly what the code meant were not.

    dwg at the start nothing about the format at all was documented. .cwp is in the same class.

    superchargers = supercharger plugs for electric cars
    This is mostly going to be pointless. Lithium ion batteries are down right expensive and heavy and slow.

    kurkosdr battery based electric cars if you add in the cost of making the battery the CO2 produced would have been less to run a petrol car.

    http://en.wikipedia.org/wiki/Hydrogen_station Hydrogen already has standard port combined with fuel cell tech far more sane than battery powered electrics.

    Sometimes the correct answer is not to standardise particular things.

  75. dougman says:

    Have to agree with KUKU on this, “Or reduce the protection to some duration that is compatible with the pace technology moves forward today.”.

    Here is a great side-article on my point: http://www.forbes.com/sites/ryanmac/2015/03/24/amazon-hammers-faa-for-lack-of-impetus-over-progressive-drone-policy/

    “Nowhere outside of the United States have we been required to wait more than one or two months to begin testing,” he said. That was supported by Senator Cory Booker, who passionately suggested that if the FAA been around during the time of the Wright brothers, other countries would have had commercial planes flying before a U.S. aircraft got off the ground.

    “This is what is hard for me to believe,” Booker said. “The slowness at which this country is moving.”

    …. working on a bill that would give businesses the right to use drones until the FAA settles on final rules in a process that could take more than two years.

    Among those companies is Amazon, which hails from Cantwell’s home state, and only last week received an airworthiness certificate to test its delivery drones outdoors in the U.S. It was a nice gesture from the FAA, said Misener, however, it did little to advance his company’s approach. It took about a year receive approval–about six months more than in other countries–and was severely limiting by only approving one model of drone to be tested.

    We innovated so rapidly that the UAS approved last week by the FAA has become obsolete,” he said. “We don’t test it anymore. We’ve moved on to more advanced designs that we already are testing abroad.”

  76. kurkosdr, recommending non-Free software, wrote, “That’s why the only alternatives to it are FOSS stuff designed to be just good enough to be used in a uni classroom, not by professionals. Same for CAD.”

    There are a lot of professionals on the planet and a lot of them do CAD but they don’t use AutoCAD. Same applies for MATLAB. Thank Goodness real professionals have choice.

    See a list of FLOSS alternatives

    As for professionals using FLOSS, check out

    • Clive Tubb, Workshop supervisor, Cambridge University, Dept. of Architecture:“At work I use QCAD 3 Pro version for all my laser cutting needs. QCAD boasts an impressive 2D toolset for drafting purposes and the modification toolset is an absolute must for me when it comes to ‘cleaning up’ student files for laser cutting.
      I am able to create new drawings and Library items and save them so easily and the library browser tool is always there to insert my blocks in a ‘snip’.”
    • QCAD manual:“QCad has an estimated user base of over 100’000 people worldwide. Its users range from industrial companies in the fields of mechanical and electronic engineering to private users, teachers and students. There are many other capable CAD systems out there and most of them offer much more than QCad does. But most of them are bound to one single platform or are simply not affordable for many potential users. That’s where QCad kicks in. QCad targets the hobbyists, occasional CAD users and people who are not CAD professionals but still need to draw plans once in a while. QCad is the CAD for the rest of us.”
    • AutoCad has only a few million installations globally.
    • The USA, alone, produces a similar number of scientist and engineers annually, people who are likely to use CAD.

    So, AutoCAD is not essential by any means.

  77. kurkosdr says:

    superchargers = supercharger plugs for electric cars

  78. kurkosdr says:

    What is really nasty here is how dwg become a defacto standard without any specifications or documentation on how it worked. The only documentation on the dwg is the reversed documentation by ODA.

    True ohioham, true. And at least AutoCAD is something used only by a certain professional sector. We have .doc, .ppt and OOXML-transitional, which are used by most of the population and are not open. Then there is .psd. And .cwp

    And governments, not having learnt from their mistake, are letting automakers impose all-kinds of proprietary, non-open standards for superchargers. When it’s too late, maybe they ‘ll take notice.

    And it’s not like “governments can’t do that”/”shouldn’t do that”. They standardized plugs and voltages, analog TV formats, digital terrestial TV formats and ASCII, and it worked quite well.

    —-

    But, the problem with your argument ohioham is that it’s irrelevant. Even if dwg was completely open, AutoCAD has so much training, documentation, and workflow habits invested into it, that escaping it is just terrible from a cost-benefit analysis.

    This is why I asked DrLoser, even if I may come out rude. I am very curious to find out what kind of project is that which needs CAD, but is not using AutoCAD.

  79. oiaohm says:

    kurkosdr depends on what you are doing if Autocad is the right tool. Remember Solidworks uses Teigha. So spending money on Autocad might have been the biggest mistake ever if the jobs in your area are Solidworks and the like.

    Files produced by Teigha open perfectly in Autocad but files produced by Autocad don’t always open perfectly in Teigha using applications.

    Problem here is Teigha is also not FOSS license compatible. But at least the parties behind Teigha have published spec sheets to dwg.

    http://en.wikipedia.org/wiki/Open_Design_Alliance

    What is really nasty here is how dwg become a defacto standard without any specifications or documentation on how it worked. The only documentation on the dwg is the reversed documentation by ODA.

  80. kurkosdr says:

    And my company, indeed the project I work on, makes extensive use of RealDWG.

    That got me curious. If I am not turning too rude, can I ask (in very loose, non-TEG-ire terms) what kind of project is that? (no names mentioned, no specifics given)

    Because, a tool like AutoCAD is something either used by a freelance professional, or by someone on a salary working for a firm. And it makes sense to buy it in both cases. It makes sense for the freelance to spend 4 grand for what is essentially the center of his business, and it makes sense for a firm to spend 4 grand per seat, in order to make sure the hired employeed sitting on the seat has proper tools to work with. Even if the firm is working on a project lasting only 3 years and has no need for CAD later, it makes sense to spend an extra 55$ per month to make sure the employee gets the job done right.

    This is also the case with MATLAB. No one seriously working with MATLAB ever thinks of an alternative, because the cost of re-acquiring documentation and training alone (let alone missing features) isn’t worth it. That’s why the only alternatives to it are FOSS stuff designed to be just good enough to be used in a uni classroom, not by professionals. Same for CAD.

    PS: In fact, if there is a good time to buy an AutoCAD license, it’s now, because it’s the last year the offer a real license, then it will be subscription-only.

  81. oiaohm says:

    Tech Soft 3D is the worldwide exclusive reseller of RealDWG. The license process may involve a thorough review of your application by Autodesk to ensure there isn’t a competition and we believe this may be the biggest first step in using RealDWG to import DWG files in your application.
    DrLoser please note is against license to make a competitor to Autocad using RealDWG. Basically I guess your developers are only building autocad extensions.

  82. oiaohm says:

    http://www.opendesign.com/the_oda_platform/Teigha
    DrLoser RealDWG is windows only. Also did you not read the bugs DrLoser.

    RealDWG .dll for 32 bit does not work on a 64 bit Windows system properly for some reason.

    Teigha is what you have to use in commercial for cross platform.

    http://prototechsolutions.com/blog/tag/realdwg/

    Yes both realdwg and Teigha suxs for different reasons.

    Note DrLoser kurkosdr question was not if commercial options existed but if the FOSS world was attempt to build their own.

    Both realdwg and Teigha licenses do a lot of forbidding.

    http://en.wikipedia.org/wiki/Open_Design_Alliance Most of the closed source competitors to Autocad use Teigha. They kind have to thinking they run on Linux and OS X.

  83. DrLoser says:

    Since RMS enunciated the principles of Free Software so many years ago, the world of IT has crossed time and space.

    I think the time for writing a script for a completely unbelievable Science Fiction movie are long past, Robert. Say, B-Movie level in the 1960s.

    I admire your total command of Time And Space, though. Now, tell us. Is parity conserved through symmetry, or is it not?

  84. DrLoser says:

    DrLoser funny skype used P2P for years until Microsoft took them over and the change become complete central server.

    Fifi funny skype was already planning to move from peer-to-peer before the acquisition. And the change is not “complete central server.” And no amount of quoting a Wikipedia article is going to change that fact.

    You just never bother to read a single cite that anybody provides, do you, you ignorant doofus?

    This proved to be a problem when not once, but twice a global Skype network outage was caused by a crashing bug in that client… bootstrapping the network back into existence afterwards was painful and lengthy, and that is in part why Skype has switched to server-based “dedicated supernodes”… nodes that we control, can handle orders of magnitudes more clients per host, are in protected data centers and up all the time, and running code that is less complex that the entire client code base. (And this conversion started well before the Microsoft acquisition was even announced, during the Silverlake era.)

    And all the other reasons why peer-to-peer was a massive scaling headache, irrespective of the takeover?

    You didn’t bother to read those, either, did you?

  85. DrLoser says:

    kurkosdr GNU is managing to make head ache for those wanting to replace Autocad.

    Interesting, Fifi. Because my company is about as cheap as it gets. And my company has a tech staff who are predisposed to anything FLOSS if at all possible. And my company, indeed the project I work on, makes extensive use of RealDWG. And there’s no sign that my company has even the faintest interest in this cobbled-together piece of FLOSS crap, upon which you are now the Foremost Technical Expert in the World, because after all you were able to google it.

    Perhaps you should have a quick fireside chat with Warren Buffet?

    I promise you — it will be quick.

  86. kurkosdr says:

    Patents, lock-up the market, innovation should be open.

    Or reduce the protection to some duration that is compatible with the pace technology moves forward today.

    That way, there would be incentive to commercialize the patent before it expires. Now, patent holders just sit on their invention, hoping that someday someone will need it bad enough to pay them whatever they want.

  87. kurkosdr wrote, “The FSF needs to focus on some “core” projects and forget about building a clone of everything.”

    For a lot of people a web-browser is just a tool to reach FB. FB is a core tool. There’s no reason GNU should limit itself to the PC or the server. The same issues that provoked RMS to start GNU are visible on the web: secret and unalterable software, algorithms, connections… If people want trustworthy and reliable software it has to be FLOSS.

  88. dougman says:

    3D printing has been around for a quite awhile, what held it up was patents! The diffusion of innovation chart for new technology states that things take time to be absorb by the populace, the phone, electric wiring for the house and even the Internet did the same.

    Patents, lock-up the market, innovation should be open. Tim Berners-Lee understood this when he created the Internet.

  89. oiaohm says:

    http://en.wikipedia.org/wiki/LibreDWG

    kurkosdr GNU is managing to make head ache for those wanting to replace Autocad.

  90. oiaohm says:

    DrLoser funny skype used P2P for years until Microsoft took them over and the change become complete central server.
    http://en.wikipedia.org/wiki/H.323

    H.323 that happens to be used being major VIOP service providers is P2P.

    Note the difference between H.323 and the old skype. Gate keeper servers run on proper servers. Yes before Microsoft took them over Skype was changing to a H.323 like configuration that would have fixed network dies just because 1 client is shot. H.323 is half way between P2p and Central Server.

    kurkosdr FSF was never coding they have always been a legal/representive body.
    http://www.freecadweb.org/
    There are open source attempts to replace cad software. There are others out side the 1 I listed..

  91. DrLoser says:

    I think this is an idea that can scale by eliminating bottlenecks. If there is no “central authority” there could be fewer bottlenecks.

    Um, that’s not actually how “scaling” works, Robert. Or rather, it is … but the bottleneck is the peer-to-peer network itself, rather than some mythical issue with a “central authority.”

    Skype, you say? They dropped peer-to-peer. It doesn’t scale.

  92. kurkosdr says:

    node Quitter.se and started posting

    What is this? Some site where cigarette addicts gather to discuss their efforts at getting rid of an addiction they shouldn’t have in the first place? (I can understand alcohol addiction, but how smokers get willingly addicted to something that stinks and tastes horribly on the first try is something I cannot understand). Quitter? That’s the best name they could come up with?

    Her Twitter followers proved willing to follow her all the way to GNU social, and began joining existing nodes en masse and starting their own.

    Which caused Twitter to lose a significant part of their audience and see a measurable drop in visits and ad-revenue… oh yeah that never happened.

    The growth was so explosive that the some of the existing GNU social nodes were unable to handle the traffic.

    Come to Mom ‘n Pop social network. Wait, wait! No more than 10 at a time.

    Thousands of Spaniards

    Thousands? So, a medium-sized village? World, take notice.

    ——

    Okay, on a more serious note, Facebook has a lock-in that cannot be easily broken. It’s what the telephone used to be. Everyone has a number (account), so it’s the first thing you ask from your friends. What’s App is also big, because it didn’t try to build a “-killer” service. It was different. It’s intent was to supplement not replace a similar product. For the billionth time, “-killer” products suck, unless they are siginificantly different or much better than the product they are destined to “kill”.

    And on a more general note, the FSF (GNU) has a serious relevance problem. The idea of “lets build a FOSS clone of every tool users need” was applicable back in the 70s and the 80s. If you “workstation” didn’t ship with a particular tool (say a compiler) you just downloaded the GNU clone. But now, does FSF really try to build usable clones of AutoCAD and Facebook?

    The FSF needs to focus on some “core” projects and forget about building a clone of everything. But of course, herding FOSS developers to a common goal is like herding a pack of cats. With some cats being narrow minded and/or really arrogant.

  93. LinuxGentlesir wrote some thoughtful comments including, “3D printing as a counter to central control over production”

    IT in general can distribute control. 3D printing could in fact be used by corporations to distribute products more cheaply and faster than freight. Manufacturing, at least in plastics, could evolve into a copying process just like publishing. Eventually this might generalize into a generic factory in every region that cranks out product X, Y, or Z according to the specs of some designer and delivered locally to some consumer. Every business might have such a printer to make products magically appear.

    e.g. My tractor was built in China using raw materials shipped from elsewhere and then shipped to me by truck, ship, truck and more trucks. Total cost of all that shipping was probably more than half the cost of the tractor in China. If robotic factories could crank out any product anywhere, the information required to do that could travel on the network and energy and raw material would only have to go in bulk to those factories. The consumption of resources and the total cost of production would be greatly reduced. It might be the only way that The West can compete with China for the next decade or so.

    3D printing can do this now for certain products just as PDF can do this now for documents. I think once the costs and reliability of 3D printing become manageable other forms of distribution could result. Except for food, this should be feasible in a decade or two. My tractor could have been delivered to me weeks sooner at half the cost if some robots across town assembled it rather than what I’ve had to do. e.g. A locally manufactured tractor could avoid the few days’ delay due to Customs inspection.

  94. LinuxGentlesir says:

    In the most abstract sense Free Software part of a greater multi-front technological struggle against central authority. It’s the most successful one so far, but I think in the next few decades you’ll see:

    – 3D printing as a counter to central control over production
    – Mesh networks and distributing systems as a counter to central control of the Internet
    – Cryptocurrency as a counter to central control over money
    – Blogs and social media as a counter to central control over information

    As such we’ll move to a world away with that has decreasing importance of hierarchical social structures like corporations and government, towards a world with greater individual liberty.

Leave a Reply