Follow Slashdot stories on Twitter

 



Forgot your password?
typodupeerror
Compare cell phone plans using Wirefly's innovative plan comparison tool ×
NASA Space Government ISS Operating Systems Programming Software The Almighty Buck Transportation Build Hardware Technology

New NASA Launch Control Software Late, Millions Over Budget (go.com) 205

schwit1 writes: The launch control software NASA is writing from scratch for its Space Launch System (SLS) rocket is way behind schedule and way over budget. According to ABC News, "Development of this new launch control software is now projected to exceed $207 million, 77 percent above 2012 projections. The software won't be ready until fall 2017, instead of this summer as planned, and important capabilities like automatic failure detection, are being deferred, the audit noted. The system is vital, needed to control pumps, motors, valves and other ground equipment during countdowns and launches, and to monitor data before and during liftoff. NASA decided to write its own computer code to "glue together" existing software products a decade ago -- while space shuttles still were flying and commercial shippers had yet to service the space station. Both delivery companies, SpaceX and Orbital ATK, rely on commercial software, the audit noted."

In other words, even though NASA could have simply purchased already available software that other launch companies were using successfully, the agency decided to write its own. And that decision really didn't come before the arrival of these commercial companies, because when it was made a decade ago that was exactly the time that SpaceX was beginning to build its rocket. This is simply more proof that SLS is nothing more than a pork-laden waste of money designed not to explore space but to generate non-productive jobs in congressional districts.

This discussion has been archived. No new comments can be posted.

New NASA Launch Control Software Late, Millions Over Budget

Comments Filter:
  • by Andreas . ( 2995185 ) on Tuesday March 29, 2016 @05:59AM (#51798535)
    That last few sentences were really inacceptable. Could someone edit this?
    • by Archtech ( 159117 ) on Tuesday March 29, 2016 @07:00AM (#51798705)

      That last few sentences were really inacceptable.
      Could someone edit this?

      OK.

      "Those last few sentences were really unacceptable".

      FTFY. There will be no charge.

      • That last few sentences were really inacceptable. Could someone edit this?

        OK.

        "Those last few sentences were really unacceptable".

        FTFY. There will be no charge.

        You, sir, have been banned for life as a slashdot editor. That's not the kind of quality we have come to expect nor the kind that has made slashdot famous.

    • SpaceX, solar power generation, battery technology, electric cars, anti-nuke, etc., etc. His shills are all over /.
      • Actually, musk and many of his followers are nuke fans. After all, they tend to be scientists, rather than simple GOP trolls.
        • GOP is anti nuke? Since when?

          I would say rabid environmentalists are anti-nuke, because they think the wind and solar power generation will totally produce enough power to decommission all the nuclear plants.

          • No, tomhath is the GOP troll.

            And yes, a number of the rabid environmentalists along with dems ARE anti-nukes. But, I do not consider them to be scientists or logical about.
            These ppl are not much different from the far right wing nut jobs that refuse to look logically at climate change.
            And yeah, the idea of our converting to 100% AE, which is mostly wind and solar, has to be one of the stupidest things that I have heard.

            BTW, I am hoping that with O being pro-nuke that during the lame-duck session,
            • Agreed. The main risks from nuclear power are all the old plants that should have been decommed long ago. I wish someone would put more effort towards LFTR, as that seems to be the safest possible nuclear power solution.

              • The weird thing is, that money that was set aside to handle the 'waste', is ideal for the plants to grab, install new lftr reactors from trans atomic and flible, and then process the old waste. In doing that, they cut the waste down to 5% size and make it safe in 200 years. This is a win-win-win. And yet , CONgress does nothing.
    • by mi ( 197448 )

      NASA could have simply purchased already available software that other launch companies were using successfully, the agency decided to write its own.

      Yeah, as if private enterprises have never made the same mistake...

      This is simply more proof that SLS is nothing more than a pork-laden waste of money designed not to explore space but to generate non-productive jobs in congressional districts.

      I'm no fan of government-run anything, but this is hardly "proof". The submitter's editorializing is, indeed, offensive

      • I'm thinking yesterday a DA orgasms over a company that m$ [slashdot.org] purchased. Today, it's this post. WTH going on at the editors desk?
      • by Immerman ( 2627577 ) on Tuesday March 29, 2016 @10:09AM (#51799535)

        > The folks, who wish to see "pure research" continue, can make donations to the non-profit without forcing others (via the IRS' implicit gunpoint) to do the same.

        An *excellent* idea. Lets do that right after we fund the military that way. And all the subsidies for agricultural and oil mega-corporations. NASA is such a tiny drop in the federal budget that it's barely worth mentioning. It's far more noteworthy for the visibility of its accomplishments (and failures) than the size of it's budget. And frankly, even at the same price I'd much rather involuntarily fund blue sky research than genocide and empire building.

        • So, since you want to get rid of the military, have you called the constitutional convention needed and written up the constitutional amendment?

          https://www.law.cornell.edu/co... [cornell.edu]

          We the people of the United States, in order to form a more perfect union, establish justice, insure domestic tranquility, provide for the common defense, promote the general welfare, and secure the blessings of liberty to ourselves and our posterity, do ordain and establish this Constitution for the United States of America.

          Last I checked, NASA wasn't MANDATED by the constitution.

    • by Columcille ( 88542 ) on Tuesday March 29, 2016 @08:40AM (#51799079) Homepage
      Indeed. I call bullshit on the summary's FUD. Criticizing NASA for not using the software of fledgling, unproven companies? Criticizing NASA for using custom software for their completely unique launch system? There are good reasons to criticize some of NASA's recent decisions, but this article is not among them.
      • by Geoffrey.landis ( 926948 ) on Tuesday March 29, 2016 @09:52AM (#51799417) Homepage

        Yes, the article seems to be trying to manufacture outrage out of nothing.

        It's a make-versus-buy decision. Industry does these decisions all the time. When your applications are unique, the decision tends to go toward "make your own;" when your application is something that many other people also do, the decision tends to go toward "buy the commercial product".

        Buying off the shelf comes with hidden costs unless what is available exactly meets your need-- if you need to write a new contract for every change (and since you still haven't designed the system you're launching, there will be a lot of changes needed, as you keep refining requirements) every single change is a chance for the vendor to demand large dollar payments.

        And the article's statement "why doesn't NASA just use what Space-X used" is absurd. Ten years ago, Space-X was an unknown company who had just launched their first rocket. Which failed. As did their next launch. And the one after that.

      • Indeed. I call bullshit on the summary's FUD. Criticizing NASA for not using the software of fledgling, unproven companies? Criticizing NASA for using custom software for their completely unique launch system? There are good reasons to criticize some of NASA's recent decisions, but this article is not among them.

        Private industry always does better, that's why they don't have anything to do with NASA. That's why they have all their own launch facilities, tracking and all aspects of operations.

        All the proof we need.

        Oh....... wait.....

      • Besides, we have to keep all of those ADA programmers employed!

    • by Anonymous Coward

      It looks like an interesting, useful, marginally informative post with an over the top conclusion.

      Let's look at that closer,

      SENTENCE N-2 :
      "In other words, even though NASA could have simply purchased already available software that other launch companies were using successfully, the agency decided to write its own." That's pretty much what the news article said, but at /. I would have hoped for a bit more details on how the actual 2 projects were done and how they turned out. Probably would have made NAS

      • As for adjusting the /. posting, there is a fine line between editorial comment and starting a PC police here, so the article should stand unless the poster wishes to adjust it. Either way seems fine. (I think maybe the cat was already out of the bag that this group likes X better than N.;-)

        I'm not so certain the line is all that fine. Let us take an example

        Suppose someone wants to post an article using one of the last gasps of denialism, the discrepancies between satellite and radiosonde data. Maybe with links to one of the denialist blogs. People such as myself will squawk long and loud to note the data is outdated, and the discrepancies have long since been correlated.

        So in fact, I like the postings of the useful idiots - they are also useful to me.

        If some useful idiot like schwit1,

  • Bad management. (Score:5, Insightful)

    by Qbertino ( 265505 ) <moiraNO@SPAMmodparlor.com> on Tuesday March 29, 2016 @06:02AM (#51798549)

    There is no software on the planet that is more scrutinised and more meticulously developed than software for spacecraft [fastcompany.com].
    Start a Softwareproject like that without having it properly planned or the right people involved and your project will go over budget manifold inmediately.
    No surprise here.

    • by K. S. Kyosuke ( 729550 ) on Tuesday March 29, 2016 @06:25AM (#51798611)

      and your project will go over budget manifold

      I am familiar with fuel manifolds and oxidizer manifolds, but where exactly is the budget manifold in a rocket?

      • by invid ( 163714 )
        Let's do the math, given a budget manifold of a hundred square meters, how fast would the velocity of cash have to be to equal the US yearly defense budget in real time, and what type of material would have to be used to withstand the heat of friction, given the cash is in one dollar bills?
      • by hey! ( 33014 )

        I am familiar with fuel manifolds and oxidizer manifolds, but where exactly is the budget manifold in a rocket?

        It starts at the House Appropriations Committee offices, runs in a zigzag path to 330 E Street SW (NASA headquarters). From there it branches out to myriad contractors strategically scattered across a majority of Congressional districts.

      • Typical engineer. Should have spent at least some of that college time in Liberal Arts classes, then you'd know that "Manifold" can mean 'many or multiple' as well.

        It's short for many-fold, ie multiple.

        • That was actually a joke. I happen to have spent a better part of a decade by translating technical English for a living.
      • by eth1 ( 94901 )

        and your project will go over budget manifold

        I am familiar with fuel manifolds and oxidizer manifolds, but where exactly is the budget manifold in a rocket?

        It's usually referred to as "accounts payable." It's an adapter that allows numerous tubes going to government contractors to connect to a single government teat.

      • Manifold:
        A manifold is a topological space that is locally Euclidean (i.e., around every point, there is a neighborhood that is topologically the same as the open unit ball in R^n). To illustrate this idea, consider the ancient belief that the Earth was flat as contrasted with the modern evidence that it is round. The discrepancy arises essentially from the fact that on the small scales that we see, the Earth does indeed look flat. In general, any object that is nearly "flat" on small scales is a manifold,

    • by MrKaos ( 858439 )

      You write software to fit the need. Nasa invented the "Agile" approach with Gemini and Mercury, but I doubt any of those guys would have to argue with their management about the need for descope in the middle of an iteration. Software is a malleable form of engineering and a number of approaches can be bought to bare to make it fit for purpose. Your appetite for risk and the impact of that risk is part of the process that decides what that is. After all how many software projects even have risk and issue lo

    • "There is no software on the planet that is more scrutinised and more meticulously developed than software for spacecraft"

      There's at least one. Software for US nuclear weapons systems. I once watched a USAF nuclear safety audit over the course of a few years. I was thoroughly impressed with the quality of the work. (Not that I thought that particular nuclear weapons system made a damn bit of tactical sense. Thankfully, it's long gone).

      BTW, I think that any software involving digital communications dese

      • There is no software on the planet that is more scrutinised and more meticulously developed than software for spacecraft

        There's at least one. Software for US nuclear weapons systems. I once watched a USAF nuclear safety audit over the course of a few years. I was thoroughly impressed with the quality of the work.

        ...and then, as an added assurance to make sure that nothing would slow down the ability to launch missiles, the code to launch missiles was set to all zeros, and never changed.

        http://arstechnica.com/tech-po... [arstechnica.com]
        http://www.theguardian.com/wor... [theguardian.com]
        http://foreignpolicy.com/2014/... [foreignpolicy.com]

    • He says at the bottom that he is a native German speaker, and please excuse his English.
    • by Lumpy ( 12016 )

      Yet during the shuttle years the mostly used software was a linux PC that translated all the older systems data streams into something easier to parse by the staff.

      Not everything is heavily tested before deployment. They even allow Microsoft OS up there for the mission report laptops and the experimentation control laptops that have to run labview.

      I wonder if the ISS laptops running windows 7 are bugging them to upgrade to 10?

    • Because if this:

      https://en.wikipedia.org/wiki/... [wikipedia.org]

      Happened at a nuclear power station or a nuclear missles launch system a lot of people would be in deep radioactive shit.

  • What kind of summary/ story is this? This was only the executive summary. We want to know what systems the stuff runs on, what programming languages are used; you know, the geek stuff!
    • by Dutch Gun ( 899105 ) on Tuesday March 29, 2016 @06:08AM (#51798561)

      The summary is happy to draw a conclusion for us though, it seems. No need to think for ourselves.

    • Yeah, missing information about the technical side of the project but also missing any details about why the project is so delayed. Is it a specific system integration causing the problem, or quality issues showing up in testing or feature creep.

      No useful information included in the summary or the original article.

    • Re: (Score:2, Informative)

      by Anonymous Coward

      https://oig.nasa.gov/audits/reports/FY16/IG-16-015.pdf

      Note that NASA is _not_ reinventing the wheel, and they _are_ using COTS, but since Commercial Software just simply isn't up to snuff off the shelf, they have to write "Glueware". Would _anybody_ trust Microsoft with this task?
      Also note that references to cost overruns includes all the work that NASA put into the Constellation Program... which was cancelled, (Well, no longer funded, after 2011.)

  • by Gravis Zero ( 934156 ) on Tuesday March 29, 2016 @06:29AM (#51798623)

    even though NASA could have simply purchased already available software that other launch companies were using successfully, the agency decided to write its own.

    NASA decided to use existing software that was known to work and was fully understood rather than rely on commercial software which could be total shit. besides, they wouldn't be purchasing the software, they would be licensing it which means they would likely have to pay $X for Y computers for Z years. also, what happens when they want to add feature XYZ and they are unable to? freedom isn't free... it's 207 million dollars.

    • Freedom _would_ be free if there were a GNU/Launch. Let's show them how it's done! Who is with me?

    • Commercial software is available now. Not so much 10 years ago.

      From the report:

      NASA made its decision regarding the architecture of the SCCS software nearly a decade ago and has continued on that path. In our view, this may no longer be the most prudent course of action given the significant advances in commercial command and control software over the last 10 years. Specifically, command and control software technology has matured to the point where COTS products may provide much of the functionality needed to launch the SLS and Orion with relatively little modification. Indeed, the two companies under contract with NASA to deliver supplies to the International Space Station - Orbital Sciences Corporation and Space Exploration Technologies - both use COTS products to accomplish their missions.

      • Commercial software is available now. Not so much 10 years ago.

        But according to TFA "even though NASA could have simply purchased already available software that other launch companies were using successfully, the agency decided to write its own."

        Are you suggesting that the submitter is confused, or perhaps being economical with the truth?

        • by Overzeetop ( 214511 ) on Tuesday March 29, 2016 @10:08AM (#51799517) Journal

          No, the submitter has 3rd grade reading comprehension skills and an agenda. And it's in the summary which suggests that software was available, TFA quote is actually "Commercial software products would be a better option for NASA as well, according to the audit, especially given recent advances in the area." Note the use of future tense, not past unreal conditional. IOW, it still might be more economical at this point, but it wasn't an option when the project was started.

          The commercial firms which are servicing the space program had never delivered a payload to orbit at the time of the original design specification and plan, and had no tested software at the time. NASA, otoh, had subroutines already written which had been tested and vetted for decades, over hundreds, if not thousands, of successful launches and NASA wanted to use those [tested] routines in their new launch control system. The contract was to put them all together into a cohesive whole. Not a trivial task, but also not writing from scratch.

            More importantly there is no COTS software in this arena. This is not Word with a custom skin. This is piece by piece built based on the unique hardware and control systems which are part of the critical safety path of the launch sequence. Even if SpaceX or another space transport company has software they use, it would have to be stripped, rebuilt, and re-tested for the configuration at the KSC launch complex to be used for these flights.

    • Re: (Score:3, Informative)

      by phayes ( 202222 )

      Launch control software is NOT commercially available for STS! Each launcher has it's own LCS because even more so that with PCs the software needs to be adapted to the launch Hardware. Ponder the hardware differences between homogeneous Falcon-9 and STS's heterogeneous mix of solid boosters and cryogenic engines. Yes, one could adapt one LCS from one launcher to another but the undertaking is more a general rewrite due to differing hardware than it is a simple recompile.

      I'm not even sure that if Nasa were

      • by Lumpy ( 12016 )

        "not Joe's Space junkyard and spare parts."

        If that place existed, I would SO spend a lot of money there.....

        "Yes I DO need a orbital insertion motor for a satellite... stop arguing and help me load it in the minivan."

  • by Salgak1 ( 20136 ) <salgak&speakeasy,net> on Tuesday March 29, 2016 @06:33AM (#51798635) Homepage

    . . . .and when the FBI started to develop its' own case manager, the "Virtual Case File [ieee.org]", which was one of the more spectacular failures in Government IT Development [cnn.com].

    When the post-mortem finally comes in, I'd be more than willing to bet that it was due to (1) lack of formalized baseline requirements to hang an initial design on, and the real program-killer, (2) constant requirements creep. Because contractors are unwilling to tell a Federal Customer "no" (because it usually results in decreases in funding in the next task order, or re-allocation of slots to another contractor. . .), there's a constant "just add this one little thing". Over and over again, until you have an unworkable mess and a design that looks nothing like the initial requirement.

    The same kind of pressures destroyed the Navy A-12 "Avenger" attack jet [wikipedia.org] in 1991: constant scope creep, until the aircraft was too heavy to fly off an aircraft carrier. The resulting legal fight lasted 13 years [reuters.com]. . .

    • . . . .and when the FBI started to develop its' own case manager, the "Virtual Case File [ieee.org]", which was one of the more spectacular failures in Government IT Development [cnn.com].

      When the post-mortem finally comes in, I'd be more than willing to bet that it was due to (1) lack of formalized baseline requirements to hang an initial design on, and the real program-killer, (2) constant requirements creep. Because contractors are unwilling to tell a Federal Customer "no" (because it usually results in decreases in funding in the next task order, or re-allocation of slots to another contractor. . .), there's a constant "just add this one little thing". Over and over again, until you have an unworkable mess and a design that looks nothing like the initial requirement.

      The same kind of pressures destroyed the Navy A-12 "Avenger" attack jet [wikipedia.org] in 1991: constant scope creep, until the aircraft was too heavy to fly off an aircraft carrier. The resulting legal fight lasted 13 years [reuters.com]. . .

      Then they aren't handling their customer properly. I used to do contracting with the Department of Defense. I also helped with business development (including some pretty huge dollar value contracts) by writing technical approaches to these RFPs. When we got these contracts I was the lead engineer and often handled some of the project management aspects as they related to the engineering efforts. I went to every meeting with the customer, from cradle to grave. And you're right, I never did tell the cus

  • Link to the report (Score:5, Informative)

    by hackertourist ( 2202674 ) <hackertourist.xmsnet@nl> on Tuesday March 29, 2016 @06:44AM (#51798669)

    28-page PDF [nasa.gov]

  • by Anonymous Coward

    After 8 dreary years of the Obama space program; the broken promises, spectacular failures and cronyism of commercial space; tedious and trivial grind of the ISS mission; the side show spectacle of billionaires increasingly absurd attempts to recover rocket stages, you'd think people would finally be ready for the successor to Apollo. Give me SLS and a real goal, to return to the moon, any day.

  • That is the whole problem with NASA, they don't kill anyone. If they at least blew up some brown children now and again this would have gotten so much funding it would have been operational already.

  • A government program, years late and millions over budget and still isn't as good as what the private sector already had? SHOCKED! Shocked I am!

    Clearly these same type of people need to be placed in charge of our healthcare!

  • by Overzeetop ( 214511 ) on Tuesday March 29, 2016 @07:43AM (#51798829) Journal

    The "outsource everything" mantra of the 80s is still with us. It doesn't say in the article, but most of the work done by NASA is actually performed by outside contractors, and wI would bet a dollar that this is the work of USA (United Space Alliance - aka Lockheed and Martin Marietta) or one of the other giant government contractors like CSC or Booz. It may be NASAs project, but congress pretty much has gutted the real workforce so everything merely has project managers rather than actual engineers.

    As for the submitter's (and, to some extent, the article writer's) take, I think they got it backwards. This project was started BEFORE there was any commercial launches of significance and so the code simply didn't exist for a robust launch control system as envisioned. The article does point out that there is more software available today, and that it could be an option. OTOH, we're talking about proprietary code from one of two competing firms with no outside review of the codebase. That's fine for putting up a couple of tons of food and electronics, but a private company has yet to successfully, reliably put humans into earth orbit. And that kind of responsibility is an order of magnitude higher than supplies.

  • ... as every popular language today states it's used by NASA!
  • A couple of key points: 1) Software development at NASA is unlike anything 99.999% of the tech heads here would recognize. The scrutiny and the level of detail and failover protection in their code is unbelievable. Lives are at stake with their code or at worst millions of dollars of hardware crash into the surface of Mars and an entire mission is ruined. For most of us, if we have a bug in our code we patch and life moves on. If NASA has a bug in their code someone may die. 2) I have a friend that wo
  • Let's see - NASA or Space-X...not even a contest
  • by Anonymous Coward

    The vast majority of software development projects are late and well over budget. There is a lot of competitive pressure that forces bidders to give false expectations on both fronts, over and over and over again. If your offer is projected to take x million and y months, and your competitors' is 4x/5 million and 4y/5 months, who is going to win the bid? Of course, the winner will probably end up requiring 2x million and 2y months to get the job completed (if at all) so next time your bid will be 3x/5 milli

  • May I perhaps suggest... a 2230 E Rocket Launch Controller?

  • ...even though NASA could have simply purchased already available software that other launch companies were using successfully, the agency decided to write its own."

    Yeah, other, incorrect words. "...when it was made a decade ago that was exactly the time that SpaceX was beginning to build its rocket." Because when a commercial rocket is being built is the time to decide it's software is ironclad? And that the software you have available from NASA that was tried and true was definitely defunct? NASA has a lo

    • Of course, captain hindsight would have been correct about this project, and I just now realized that after making my post. Thanks captain hindsight!

  • There is no Commercial Off-The-Shelf launch software.
  • SLS may be really The Crap. However, the original post is surely the proof Space is too sweet for the flies to leave the pie in peace...

    Whoever thinks "private" soft is more effective than any government funded project is dumber the dumbest ass on the road. He does not know nothing about space travel, specially the fact that there are still lots of unknowns behind the "simplicity" of celestial mechanics. These problems are hard enough to turn a well-planned trajectory into the nightmare of a very fluid situ

  • I don't know for sure, but I would bet that it involves government contractors. Have you ever looked at how they actually choose contracting companies?

    It starts with the fact that they have all kinds of preferential treatment involved. Is your company run by veterans? How about a disabled veteran? How about a disabled veteran woman? Best yet, a disabled veteran native American woman! Now you get the contract for sure!

    I interviewed at a few of these kinds of shops. The heads were all FIGURE heads
  • Last paragraph is Pathetic Click-Bait by some clueless Internet Dude who sniff's and licks Elon Musk's ass. "Ain't nobody got time for that." Off to a real news site.
  • I'd like to know what programming languages, operating systems, development methodologies, quality assurance, etc. are being used. COTS software is usually not designed to be used in the critical operations which NASA performs. Lots of software even says that it is not for such things in the license agreement.

    We have to keep the typical moron programmer far away from NASA projects. It is customary for programmers to criticize MBAs and other managers for having no coding knowledge. Well, too often coders hav

  • by Indigo ( 2453 )

    Slashdot posts have had their ups and downs over the years, but this kind of information-free, foaming-at-the-mouth crap is really beyond the pale. The article submissions should include at least some informative content - if any dead strawmen need further beatings, why, that's what the readers are here for.

    As to the topic at hand. Love the SLS or hate it, whoever thought NASA should have used off-the-shelf software for SLS launch control should be punished by 5 years working SAP ERP's support desk. Well, m

The computing field is always in need of new cliches. -- Alan Perlis

Working...