Slashdot Mirror


Red Hat CEO Predicts Open Source Infrastructures With Proprietary Business Functionality (fortune.com)

An anonymous reader summarizes the highlights of Fortune's new interview with Red Hat CEO James Whitehurst: A recruiter told Whitehurst the culture at Red Hat was "a little bit like that Blues Brothers movie, when Dan Aykroyd says, 'We're on a mission from God.'" But Whitehurst says geeky passion "makes it a great place to be a part of," and even argues that the success of Microsoft in the 1990s can be attributed to its Microsoft Developer Network, which led developers into Microsoft's platform and infrastructure. "Developers now are heavily using open-source tools and technology and, bluntly, I think that's why Microsoft had to open source .NET and why they're embracing more open source in general. Because open source is where innovation is coming from and is what developers are consuming, it forces vendors to participate."

Looking towards the future, Whitehurst says "A rough line would be almost to say most infrastructure is going to be open source and most business functionality above it is going to be proprietary." And he also warns open source companies, "if you don't have the unique business model that allows you to add value on top of the free functionality, in the end you're going to fail... a lot of open source companies have come and gone because they've been more focused on the functionality versus how they add value around the functionality."

53 comments

  1. Reading the headline... by Anonymous Coward · · Score: 0

    ...it sounds like the CEO is predicting his own company's business model since 10+ years?

    1. Re:Reading the headline... by Narcocide · · Score: 1

      That's the first thing I thought too. It reads like RedHat is claiming to have invented some new revolutionary open-source business strategy to quickly cover up the fact they just figured out what everyone else already knew.

    2. Re:Reading the headline... by mykepredko · · Score: 2

      I haven't been around the software entrepreneurial scene for a few years but what resonated with me was that Whitehurst noted that open source (actually any company) must be able to provide a return on investment and not just value to customers/society at large.

      When RIM was crashing, I saw a number of ex-employees pitch and get investments in open source based applications which did do things that provided significant value to customers but there wasn't a clear case that anybody would pay for the end product. Many of these products used the "freemium" model in which the base functionality was good enough for customers to use without having to take the plunge and actually pay for the product.

      I know the money people considered themselves smarter than the average bear, but they really didn't go in understanding what Whitehurst said in TFA and ended up losing their investments, painfully, over a few years.

    3. Re:Reading the headline... by Anonymous Coward · · Score: 0

      Next up, RedHat invents internet to interconnect everyone.

    4. Re:Reading the headline... by Anonymous Coward · · Score: 0

      Because they have systemd now

    5. Re:Reading the headline... by Anonymous Coward · · Score: 0

      Oh sorry, I ddin't know that only RedHat used systemd... They don't so the constant moaning about it is getting rather tiresome.

    6. Re:Reading the headline... by Etcetera · · Score: 1

      Oh sorry, I ddin't know that only RedHat used systemd... They don't so the constant moaning about it is getting rather tiresome.

      They aren't, but RH corporate inexplicably pushed it despite greybeards thinking moving wholesale to an unproven system with a leader with known issues was probably a bad idea.

      Ironically, systemd solved problems that were mostly already-solved in RH-land, which is the big reason for the pushback ~2014 when it finally hit EL7 and enterprise admins had to actually care about it. (Boot speed? Please. Could have gotten a lot of that by mimicing Debian's use of DashAsBinSh. Virtually everything else other than cgroup management already had better discrete tools for management in the ecosystem.)

      I tend to think systemd's adoption was just a classic case of organizational disaster, pushed by a FreeDesktop team with an agenda and myopia, and project managers with more faith in developers than the sysadmins who actually run the product. But "proprietary complexity on top of open source" is another explanation, given how simple-to-grok shell scripts were replaced with a technically-OSS 100K LoC mishmash of non-deterministic spaghetti.

    7. Re:Reading the headline... by Kjella · · Score: 1

      Many of these products used the "freemium" model in which the base functionality was good enough for customers to use without having to take the plunge and actually pay for the product.

      Well it's tough to convince people to take the plunge on a relatively obscure product too, companies spend a lot of money essentially making sure the market knows what solutions they provide and the quality of their product. That's the point of this model which separates it from a demo or trial version, you actually get some basic functionality for free and because of that lots of people use it and lots of people have heard of you. The free part is basically advertising, it's not a bad model. It's only a bad model if you used up all your killer features on the free part because volume is everything and forgot that you need to have a conversion rate to actually make money.

      It's hard to tell how the price/quantity curve works, if you got 1000 customers willing to pay $1000 that's a million. If you got a million customers and 10% willing to pay $10 that's also a million. If you got a hundred million customers and 1% willing to pay $1 it's still a million. Are you making the most money increasing margins with a better "premium" product or increasing volume by making a better "free" product? Most startups tries for "go big or go home", if you first hit a critical mass you might suddenly be making lots of money. Most will fail though, but then most ideas do fail in general.

      --
      Live today, because you never know what tomorrow brings
    8. Re:Reading the headline... by Anonymous Coward · · Score: 0

      Used to be you didn't need a tool to manage cgroups, because it was just a bunch of fs manipulations.

      But then the cgroups maintainer (a RH employee, natch) decides that this has to change, and that a daemon has to do it instead. And naturally systemd is quick to support that thinking...

  2. enclosure of the commons by Anonymous Coward · · Score: 0

    another freeloader trying to make a buck.

    He and redhat are not on our side, tries to re-invent corporate unixes with lockin and excess profit. But that's not why we chose linux and Free as in freedom.

    Enforcing his vision via systemd.

    1. Re:enclosure of the commons by HanzoSpam · · Score: 1

      At least they're creating a great opportunity for a linux vendor that doesn't use systemd.
      I wonder how long before we have a reprise of the unix wars?

      --

      Progressivism: Parasites helping parasites to help themselves - to other people's stuff.
    2. Re:enclosure of the commons by sinij · · Score: 1

      Why are you still complaining about RHEL? Just use CentOS.

    3. Re:enclosure of the commons by Anonymous Coward · · Score: 0

      > another freeloader trying to make a buck.

      I get your point, and still... it's unfair. What Redhat is doing in terms of financing (really) free projects goes far beyond that what others are doing.

      We, as non-customers of Redhat profit from that.

      > He and redhat are not on our side [...]

      If I get your vision right (what's your vision, anyway?) society has to change by quite a bit -- one small company can only do one small thing. What have you done lately to bring this change about?

      > Enforcing his vision via systemd.

      This just killed your credibility. Mind you, I don't like systemd either, but that's just useless flamebait. If you really care, go do something: stop blathering.

    4. Re:enclosure of the commons by Narcocide · · Score: 1

      AFAIK CentOS uses systemd now too.

    5. Re:enclosure of the commons by Anonymous Coward · · Score: 0

      We don't even need to go with Linux. Solaris is usable at no charge, has an actual working enterprise filesystem, and has been around far longer than Linux has. Solaris also doesn't have systemd, and is quite stable.

    6. Re:enclosure of the commons by Anonymous Coward · · Score: 0

      If only the solaris shell wasn't such a quirky and broken POS. I'll stick with the BSDs.

    7. Re:enclosure of the commons by Anonymous Coward · · Score: 0

      The core of CentOS is just a repackaged RHEL, they don't make any actual decisions on their own. Then they have some extras on top of that, but the point really is CentOS makes no decisions about this on their own.

    8. Re:enclosure of the commons by Anonymous Coward · · Score: 0

      There is NOTHING to stop you from taking the sources to RHEL and creating your own distro without systemd.
      are you going to do it? Come on please tell us, I am sure that there are a lot of systemd haters out there who'd love to pay you for your time and effort.
      So, are you gonna do it?

    9. Re: enclosure of the commons by Anonymous Coward · · Score: 0

      I didn't know unix systems can run only one shell...

    10. Re:enclosure of the commons by hitmark · · Score: 1

      Already there:

      RHEL vs Oracle vs MS...

      --
      comment first, facts later. http://chem.tufts.edu/AnswersInScience/RelativityofWrong.htm
    11. Re:enclosure of the commons by Anonymous Coward · · Score: 0

      Bingo. RH used to publish patch sets that allowed others to rebuild RHEL without the branding and support subscription. CentOS was the best known such repackaging. But then Oracle, a long time collaborator with RH decided to do the same, and market it Oracle Linux. On top of that Oracle offered existing RHEL support customers a sweet deal where they could consolidate their expenses with a single Oracle support contract. This lead to RH changing their policies to make it harder for others to repackage RHEL. While this was mainly targeted at Oracle, CentOS got caught in the crossfire. As a fig leaf, RH then took CentOS under its wing.

    12. Re:enclosure of the commons by Anonymous Coward · · Score: 0

      Given that RH has control over Gnome, Freedesktop, and a bunch of other Linux user space plumbing thanks to employing the maintainers, you may as well start from Linux From Scratch and work your way up.

      It is quite telling that Torvalds didn't accept a job offer from RH back in the day, instead trying to stay independent. Because thanks to that Linux the kernel shows up in many places, while GNU/Linux (or these days XDG/Linux?) has languished.

    13. Re:enclosure of the commons by Anonymous Coward · · Score: 0

      Gnome, freedesktop, systemd, gtk, flatpak. It seems to me that Red Hat is trying to do a massive power grab on the linux ecosystem (at least on the desktop) by trying to uniformize every distrib through the solutions it controls.

      And the scariest thing is that this uniformization of linux on the desktop (which in itself is dangerous imho) is making linux more and more windows like (registry like configuration manager in gnome, systemd being a kernel on top of the kernel, gnome and gtk shedding functionalities left and right), flatpak being basically the windows way to install stuff, etc.

      And most major distributions happily follow. We should remember, and that article suppoting it, that RedHat is a for profit company whose purpose is to make money. What is good for Red Hat isn't necessarily what is good for the gnu/linux community.

  3. cs on veins by Anonymous Coward · · Score: 0

    who is he? a computer? if it's human, I'm not impressed when doing predictions.

  4. Words Matter by dlb101010 · · Score: 1, Troll

    that's why Microsoft had to open source .NET and why they're embracing more open source in general.

    (Emphasis added.)

    Yikes. Did the CEO of Red Hat really want to use that particular word?

  5. translation by Gravis+Zero · · Score: 1

    "if you don't have the unique business model that allows you to extract money from users on top of the free functionality, in the end you're going to fail... a lot of open source companies have come and gone because they've been more focused on the functionality versus how extract money from users."

    Red Hat has managed this is by replacing things that worked with "better" versions that mostly worked, so you would pay for their support for when it breaks.

    --
    Anons need not reply. Questions end with a question mark.
  6. Systemd - by Anonymous Coward · · Score: 0

    The art of converting the Linux ecosystem into a shittier version of Windows. And one day when something better than systemd comes out, or the inevitable need for total re-haul occurs, the whole ecosystem will fall into ruin when systemd starts being an excuse unto itself for why it shouldn't be replaced due to over-dependence.
    Can't wait.

  7. One reason I see a lot of open source by rsilvergun · · Score: 1

    is it's free, so that's what the India tech workers get trained on. Training in India is cutthroat and cheap so they're not going to pay for software unless they have to. That's also why being an Oracle DB is one of the few things that's been a sorta safe haven for workers in the America, Canada & the UK. It's too expensive to train our replacements.

    There was an article on folks switching from Oracle DB to Mongo DB on /. the other day and I suspect that's what's really driving the change. Like I always say: Good enough is always good enough.

    --
    Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
    1. Re:One reason I see a lot of open source by unixisc · · Score: 1

      But there are far fewer Linux/Unix programmers from India than there are Windows programmers

  8. more idiots by Anonymous Coward · · Score: 0

    so you would pay for their support for when it breaks.

    which you can purchase from ANYONE because it's all open source, you can even support it YOURSELF if you want

    1. Re:more idiots by Anonymous Coward · · Score: 0

      Yes and no.

      This was perhaps the case when everything was shell and daemons.

      But these days you have the byzantine blob of systemd and dbus.

  9. Open Source doesn't care for your software freedom by jbn-o · · Score: 5, Informative

    Let it never be said again that there's no substantive difference between free software and open source—here you have an open source booster (Red Hat's CEO Jim Whitehurst) pitching proprietary software as a good thing unto itself. Many years ago the Free Software Foundation told us about this when they wrote about the "Fear of Freedom" and the section that highlights how open source enthusiasts and free software activists react radically differently to non-free software:

    The idea of open source is that allowing users to change and redistribute the software will make it more powerful and reliable. But this is not guaranteed. Developers of proprietary software are not necessarily incompetent. Sometimes they produce a program that is powerful and reliable, even though it does not respect the users' freedom. Free software activists and open source enthusiasts will react very differently to that.

    A pure open source enthusiast, one that is not at all influenced by the ideals of free software, will say, "I am surprised you were able to make the program work so well without using our development model, but you did. How can I get a copy?" This attitude will reward schemes that take away our freedom, leading to its loss.

    The free software activist will say, "Your program is very attractive, but I value my freedom more. So I reject your program. I will get my work done some other way, and support a project to develop a free replacement." If we value our freedom, we can act to maintain and defend it.

    Whitehurst mentioned "why Microsoft had to open source .NET". What freedoms does that really convey to .NET users? It's worth taking a look at Microsoft's Patent Promise for .NET Libraries and Runtime Components and understanding its limitations. This patent promise doesn't look out for your software freedom. As End Software Patents warned us two years ago:

    [Y]ou're only protected if you're distributing the code "as part of either a .NET Runtime or as part of any application designed to run on a .NET Runtime". So if you add any of the code to another project, then you lose protection and MS reserves the right to use their patents against you.

    Secondly, the protection only applies to a "compliant implementation" of .NET. So if you want to remove some parts and make a streamlined framework for embedded devices, then your implementation won't be compliant and the protection doesn't apply to you.

    Microsoft's "patent promise" so-called "protection" looks very different from how the GPLv3 treats users. End Software Patents summarizes the GPLv3's language in section 11: "[c]ode distributed under the GNU GPLv3[] comes with a patent grant which basically says the contributors can't use their patents against the users for exercising the freedoms granted in the licence" whereas Microsoft's "protections disappear very quickly for those who wish to modify or re-use the code".

  10. Welcome to tools by holophrastic · · Score: 4, Interesting

    The hammer is open source. So is the screw driver, nail gun, wrench, and plyers. Tools have always been open source. Tools have never been for the end-user. The end-user has needs and requirements, and isn't interested in building it themselves. That's where the expertise of having-done-it-before is valuable. That's why we pay people to do things that they've been doing for others for decades. Of course I can learn to do it myself. I can learn to do anything that millions of men have learned to do before me. But I'm not interested in sewing my own pants.

    I'm not even interested in repairing the stitching in one inch of my pants.

    And yet, the needle and thread, sewing machines, and wood-working tools are all open source.

    Do I build my own couch? I could. It's really easy to cut wood, screw it together, cover it with foam, cover it with cloth. It's really easy to follow a pattern and a design and a template. Still, no thanks, not interested.

    I pay for someone else to build my couch because I'd rather spend my time working in my chosen profession than building a couch.

    Open source doesn't change anything to the end-user. My clients who sell white tube socks aren't going to build their own web-site. Sure they could, but they aren't interested. They also won't be their own security guard (also open source), paint their own offices (brushes are open source), or even ship their own desks (again, open source).

    Every tool, and every obvious technique is open source. Who cares. You pay someone else to use those tools for you.

    One day, 3D printers will become ubiquitous. And still, it won't matter. I'll want a widget this big and this shape to do this -- and I'll pay someone to design it. Whether they cut it out of wood, or mold it out of plastic, or hit print, is totally meaningless to me. I don't care what tools they use. I want my widget. And no, I don't want someone else's widget. Their widget won't fit my business model.

    1. Re: Welcome to tools by Anonymous Coward · · Score: 0

      None of those tools are actually free. You get what you get when you buy one, and you do have to buy them. Not one of them come with the blueprints for reproducing them.

      There is actually competition among hammer manufacturers, because there is money to be made.

      There is little to no competition among open source tools because there is so little to gain.

    2. Re: Welcome to tools by holophrastic · · Score: 1

      You are confusing "free" with "open source". Hammers aren't free, but they are open source. You don't get the blue-prints, just like you don't get the code-flow diagrams or design commentary or code documentation. But you do get the hammer, and you can see everything that the hammer is. You can peel off the handle, and you can test the material, and you can see how it works, why it works, and when it fails.

      Hammer manufacturers compete on many things (including price), but mostly on what design elements matter most. Innovating a design element is a big deal, as is patenting it. But there's no mystery to the end-user. It is better because it's this shape, or it's buttor because it's this material. There aren't many other attributes to most tools.

    3. Re: Welcome to tools by bluefoxlucid · · Score: 1

      People generally equate "open source" in software with the concept of "free" because it is implied you're allowed to have a software without paying for it, and you can convert the source to software.

      If your friend makes you a copy of a hammer, he must expend several hours of time, plus money in materials and fuel, to produce that hammer. If your friend makes you a copy of a software, he only needs to drop it into a file share or burn it to a CD--actions which take seconds. A Web site providing copies of software can have so many downloads as to reduce replication cost to many tens if not hundreds of users per penny, with the cost representing--at a minimum--the wages paid to operators (including the infrastructure allowing you to get on the Internet).

      Intellectual property is such a big deal in software because software isn't a physical thing. Software's physical existence is in the arrangement of electrons in a material that must exist whether it contains that software or some other software; and transferring the pattern to invest that material with that information is cheap as all hell. Software isn't silicon cells holding a charge or iron-oxide coatings holding a magnetic field; it's what those states represent. It's exclusively information, and so control of who is allowed to have that information is the only control you have over software.

      Toolmakers largely want you to pay them for the tools you use to do the work somebody else pays you for. You can't make a good hammer on your own; and even if you could, you'd need a small-scale manufacturing facility to do so efficiently, and you'd have to make thousands of hammers to make them per-unit cheaper than just buying a $22 Eastwing framing hammer from Home Depot, and your hammer would be a lot shittier than anything Eastwing is willing to stick their brand on.

    4. Re: Welcome to tools by holophrastic · · Score: 1

      You're totally wrong from the start.

      First off, "FOSS" needs the "F" for a reason.

      Second, source code cannot be turned into software. I don't know why you think that it can. Some source code requires days of powerful machine time to turn into software. Most requires a compiler that likely isn't free. Many need a whole host of compiler instructions, configurations, and a full IDE-worth of resources specific to the developer's choosing to be compiled.

      Dude, I build web-sites and even my shitty perl code is useless without three days of custom linux and apache tweaking. It took me three months to configure a plain-jane lamp server -- and that's with my entire business resources behind it -- pushing for a vital and highly profitable supplier change.

    5. Re: Welcome to tools by bluefoxlucid · · Score: 1

      First off, "FOSS" needs the "F" for a reason

      In your nerd-world maybe; in the world of business executives, systems administrators, and line management, "open source" means "free shit that comes with Linux". Half of them don't understand the difference between "open source" and "freeware", and some people call Linux "freeware".

      Second, source code cannot be turned into software. I don't know why you think that it can. Some source code requires days of powerful machine time to turn into software.

      This is logically-inconsistent. It can't be done, and it takes days to do?

      Gentoo compiles from stage 1 in under a day on a modern commodity PC. If you want KDE instead of Gnome, you need 40 hours. To turn that code into software, simply run the build process and go somewhere else for 40 hours. Do other work.

      Dude, I build web-sites and even my shitty perl code is useless without three days of custom linux and apache tweaking. It took me three months to configure a plain-jane lamp server -- and that's with my entire business resources behind it -- pushing for a vital and highly profitable supplier change

      I converted Dell NetVault's install instructions into a viable Docker container after a half hour of minor reverse-engineering. Turns out it stores some important state data in some files; other files can get wiped between backups with no problem.

      I've made a lot of third-party applications run in unsupported ways inside Docker containers, using distributions that they don't support, or otherwise shoehorning it into my requirements. It typically doesn't take very long--a couple of hours, at the worst--and it's easily-replicated after the first time. Once I've figured out how, I can repeat the process in a few minutes.

      Maybe you're just bad at what you do. The core of economics is technical progress: find a way to accomplish the same goal with less work (human time investment). Much of what you describe takes me only minutes to do even faced with awkward bullshit; I've gotten good enough at probing strange software to handle the really bad cases in a (large) portion of a day, and that's hardly-ever happened. When I started doing all this crap, it would take me weeks or months to get it quite right; but I got better with time and tools.

    6. Re: Welcome to tools by holophrastic · · Score: 1

      I'm going to stop at your logically-inconsistent accusation based on your logically-inconsistent inability to read.

      I said "source code cannot be turned into software" and "source code requires days of powerful machine time to turn into software". The former is missing the requirements listed in the latter. If you can't compare two sentences, and see the big huge difference, then we're done here.

      So I'll summarize, for you, to satisfy your early complaint -- alone, without some amount of help, expertise, documentation, know-how, assistance, time, resources, and understanding, most source code cannot be turned into software merely by commonly available computer hardware.

      You can list myriad examples of source can that's been designed to do exactly that, but I'm not saying that all source code needs more, I'm saying that most/much/some/many/any does/can/might/will. When I don't use a modifier, you don't get to presume I meant "always/forever/in-all-instances/every-time".

      You are arguing based on syntax, in a debate based on abstract concepts. Given recent political events, I can guess where you learned your style. I'll guess that you also think oral sex isn't sex -- even though it's right in the name.

    7. Re: Welcome to tools by holophrastic · · Score: 1

      I decided to read the rest of your post. A few things stand out.

      First, you're talking about software application (mostly), that perform a set of tasks within a system -- so you can mimic such a system with a container, shoehorns and all. I'm talking about software married to the system. You can't shoehorn it into another system, it is the entire system.

      Second, you're now a professional. If you're arguing that a professional can build open source code into software, the only word left to say is "duh". A professional glue manufacturer can copy someone else's glue too. The difference between you and the original author, especially after a half-day of study, is almost nill. The difference between you, and the typical user of said software is all of your tools, training, and experience.

      Third, I wrote every line of code for my software. I did every design chart. I decided what it needed to be, and how it would work. I used it for years, and based my entire livelihood on it. I employed dozens of people. Then, with international issues being what they are (CIA directly connected to Rackspace), I changed data centre providers to save over $15'000/year. It still took me three months, and three persons, to set up the old software on the new network, and took me two years to plan the move.

    8. Re: Welcome to tools by bluefoxlucid · · Score: 1

      I said "source code cannot be turned into software" and "source code requires days of powerful machine time to turn into software". The former is missing the requirements listed in the latter. If you can't compare two sentences, and see the big huge difference, then we're done here.

      That's not how language works. You stated an absolute premise when you said source code cannot be turned into software, the same as if you said mere thought cannot bring matter into existence; then you said a machine can turn source code into software. Which is it? That there is no means by to turn source code into software, or that there is a means by which to turn source code into software?

      alone, without some amount of help, expertise, documentation, know-how, assistance, time, resources, and understanding, most source code cannot be turned into software merely by commonly available computer hardware.

      Categorically false. Most source code can be turned into software with less required effort than creating the source code from a requirements document.

      When I don't use a modifier, you don't get to presume I meant "always/forever/in-all-instances/every-time".

      Without a modifier, the general or vast-majority case is implied. Logically, there must be minority or specially-crafted cases where the general rule does not follow; depending on what is being discussed, this can be due to obvious practice (as with source code) or to routing around physical impossibilities (e.g. a man cannot fly--unless given a hang glider, jet pack, or other sort of device to modify the way we approach the laws of physics).

      You are, in fact, saying what any common English-language reader will interpret as "most". When I say, "Dogs don't meow," it is assumed by a reader unfamiliar with a "dog" that all or nearly-all dogs do not meow; it would be patently-absurd for the reader to assume that some minority or less-than-extreme majority of dogs do not meow based on that statement. Even legal arguments make this interpretation, based entirely on the understanding that a reasonable person will make this interpretation.

      Any linguist will tell you that languages contain loads of implied meaning. "Most or all" is the implied meaning of general statements.

      You are arguing based on syntax, in a debate based on abstract concepts.

      You're trying to hold an indefensible position by arguing that the basic interpretation of English doesn't apply to your statements. Would you like to redefine "the" or "is" as well? Perhaps we should assume the definite case ("is") is the indefinite case ("might be"). Maybe we can redefine MUST as SHOULD, while we're at it.

      Your argument is essentially that you can be wrong or flatly lie, because whatever you say has a different meaning that only you can know, and thus that you are not speaking English.

    9. Re: Welcome to tools by holophrastic · · Score: 1

      I'll stop again at your start. In order for the statement "source code cannot be turned into software" to be false, you'd need to demonstrate that all source code everywhere for all time can be. I did not state an absolute. You interpreted it thusly.

      But again, this is all meaningless. This is you arguing syntax again. Stop picking one statement of mine that has nothing to do with the argument being made.

      I argued that tools, in general, tend to be just as open source as software. That a hammer is just as open source as apache. Argue that.

    10. Re: Welcome to tools by bluefoxlucid · · Score: 1

      Software "married to the system"? You write perl code; you're not talking about kernel modules integrated with a monolyth.

      Third, I wrote every line of code for my software. I did every design chart. I decided what it needed to be, and how it would work. I used it for years, and based my entire livelihood on it. I employed dozens of people. Then, with international issues being what they are (CIA directly connected to Rackspace), I changed data centre providers to save over $15'000/year. It still took me three months, and three persons, to set up the old software on the new network, and took me two years to plan the move.

      I had a discussion about using AWS versus Azure recently. The concern raised was that moving off AWS if it became hostile would cost years of time and lots of risk in transition.

      I created a small document summarizing the advantages of deeply-integrating with AWS, as well as the risks, and the manner in which to isolate those risks. I've done the same with various other softwares to connect various tools. Modern software design is founded on concepts like encapsulation, where a particular part of a system does a particular job, and where the interfaces between these systems are borders; you can rip those borders away and create new borders as you like. In software design patterns, they use things like "bridges" and "adapters" to do this.

      Primarily, we design our systems to make all the logical decisions and track all the state they need; then, when we want to interface with some other system, we add a piece of code which can accept our decisions and our state, and can take the relevant actions. Automatic scaling, alerting, or filing information in databases, it's all worked into whatever target. We can switch out any interface at will. It's even possible to store all of our information in some back-end system like MongoDB or PostgreSQL, and then instruct the system to integrate old information with new tools or with an upgraded tool that accepts more of the full data, by which a simple strip of code runs through our back-end information and tells the glue code to find the matching information and update it--bluntly, so that the adapter decides what information the target system handles, and so will add information if that system now handles additional data, without any of the other parts of the programming knowing the details.

      Re-engineering in a deeply-integrated migration is small when done as such. It's precisely as large as writing code to talk to the new targets; every other piece is completely-naive about what's on the other end, so no code is written there. Three programmers here could switch from AWS to something completely-different and feature-complete with performance statistics, autoscaling, and S3-like storage in under a month; migrating the actual software might not take as long, since the software moves easily. The biggest problem would be planning the disruption as DNS changes and data migrations took place.

      I should know. I've done it twice.

      I've also done it in cases where we didn't have the same control. I've dealt with migrations of thousands of Web sites and a media CDN to a completely-new system, with new Web hosting, new Web site software, a new DNS provider, and so forth. We performed migrations in batches of under ten sites per week. Each batch of migrations took under an hour to complete; we moved at an artificially-slowed pace to reduce risk, allowing us to respond to any discrepancies. We retrained the content producers after each migration, as well. It took under six months in total; migration itself--from one system to something completely-different in practically every manner--took about 30 hours, plus a week to write and test the scripts to dump and load the data.

      That wasn't a matter of moving software--when I moved the existing CDN software from one system on one OS version to a completely-new system on a new OS version, it took me 4 months to build the new system

    11. Re: Welcome to tools by Anonymous Coward · · Score: 0

      "Any linguist will tell you that languages contain loads of implied meaning. "Most or all" is the implied meaning of general statements."

      Wow, I could have used your help a few weeks ago when I got clobbered on twitter for calling into question a statement " believe x". The writer claimed that of course the word "some" should be assumed. I question this sort of generality regardless if the person is conservative or liberal and everybody hates it. So I gave up on twitter.

    12. Re: Welcome to tools by Anonymous Coward · · Score: 0

      Yikes I shouldn't have used brackets. The quote should have been "people of a certain race believe x".

  11. actually... by Anonymous Coward · · Score: 0

    that's just this asshole's wishes, now that they fucked up linux with that systemd shit the shit gets deeper

    1. Re:actually... by Anonymous Coward · · Score: 0

      Oh Shut the fuck up you winger. If you don't like it, go use Windows.

  12. Software-as-a-charity by unixisc · · Score: 1

    another freeloader trying to make a buck.

    He and redhat are not on our side, tries to re-invent corporate unixes with lockin and excess profit. But that's not why we chose linux and Free as in freedom.

    Enforcing his vision via systemd.

    Since they are the only successful Linux company ever, I'd hold my horses before I bashed (no pun intended) them. Guys like RMS might, since they believe that people should just have altruistic goals and nothing more, but that's not how the real world works. Similarly, people just writing software and never expecting to get paid is something that we should accept as being rarer than pink unicorns w/ gold plated serrated horns

    1. Re:Software-as-a-charity by Anonymous Coward · · Score: 0

      Guys like RMS might, since they believe that people should just have altruistic goals and nothing more

      [[citation needed]] ... oh; you don't have one because RMS has many times supported almost anything which has profit making goals as long as it follows his proposed rules. If the foundation of your comments is bullshit then we should discard the rest before we sink in with it.

  13. Re:Open Source doesn't care for your software free by mvdwege · · Score: 2

    here you have an open source booster (Red Hat's CEO Jim Whitehurst) pitching proprietary software as a good thing unto itself.

    This does not have to mean what you imply it means. In-house business software is by its very nature proprietary. What Whitehouse is doing is essentially telling business that it is okay to build your own proprietary business software on top of FLOSS architecture, aka he's countering the usual 'if you use GPL software you must Open Source your internal software' FUD.

    --
    "I know I will be modded down for this": where's the option '-1, Asking for it'?
  14. Open Source Foundations. by Anonymous Coward · · Score: 0

    And he also warns open source companies, "if you don't have the unique business model that allows you to add value on top of the free functionality, in the end you're going to fail... a lot of open source companies have come and gone because they've been more focused on the functionality versus how they add value around the functionality."

    While this is true, one important part of open-source is that those "failures" don't go to waste. In other words someone had to build all those foundations before someone else could come along and build a house on top of it to sell.

  15. Re:Open Source doesn't care for your software free by olau · · Score: 1

    ... here you have an open source booster (Red Hat's CEO Jim Whitehurst) pitching proprietary software as a good thing unto itself.

    I think he's just commenting on the fact that while people are still deploying and using proprietary software, they're increasingly going to do it on open-source infrastructure.

    So while he's not pitching proprietary software as a bad thing, I think it's quite a stretch to claim the opposite, from this story.

  16. Re:Open Source doesn't care for your software free by bluefoxlucid · · Score: 1

    Okay so there are two take-aways here.

    One is that anything that comes out of Redhat's mouthpiece is usually self-aggrandizing bullshit. That's not to say it's always wrong; just it's not well-developed. I can say any stupid thing based on first-observations and be right 0.1% of the time--that's approximately how economic political debates work (everything in economics violates common sense with an unlubricated broom handle). When Redhat is right, it's typically because their bullshit lined up with reality by a chance meeting.

    The other is that Free Software Advocates like Stallman are die-hard communists. Stallman doesn't care about your freedoms; he wants to pry your work out of your hands on his terms. Stallman has said, directly and unapologetically, that the GPL's purpose is to force people to release their work to the world for free. He doesn't care about the labor a person puts into making a thing, and about his right to profit from his labor--or to attempt to in whatever way he sees fit, even if that way is misguided. Stallman doesn't believe in markets supplying an alternative. He believes that keeping control of your own work is theft, and that your time and effort belong to the world.

    Free software advocates as such like to dismiss this line of reasoning by placing an opaque cover over the 150,000 man-hours used to make a relatively-complex application and point at the 0.001 man-hours used to supply a download pipe to copy it, claiming that's the entire cost they owe for the product, because they're not placing any load on the creator or taking anything away from him for making a copy. Never mind that he has no way to profit from it except by begging that someone pay him out of the very goodness of their hearts; these people believe that, despite not being compelled away from copying the product for free, they wouldn't be compelled to pay any price for it in any circumstance, but that obviously someone else would pay exactly the same for it whether it was given to them for free or dangled there with an attached price tag.

    In effect, these people believe that consumers, given two prices ($0 and $50, for example), will always opt for the one the seller actually wants them to pay, or at least will actively-avoid the lowest one and pay the highest they are willing to pay in all cases.

    Once you understand that, you never have to listen to them again, because theory-of-mind allows you to estimate what someone will say about anything even when that person's thinking is horribly wrong. Eventually, you don't have to ask RMS what he would say about a thing; you only have to ask yourself what RMS would say.