Slashdot Mirror


Time To Move on from DevOps and Continuous Delivery, Says Google Advocate (zdnet.com)

A reader shares a report: Continuous improvement and continuous delivery (CI/CD) and DevOps may be on many peoples' minds these days, but there's nothing particularly new about the concept -- software shops should have put these concepts into action years ago. Instead, technology leaders should be now worrying about the futures of their businesses. That's the view of Kelsey Hightower, staff developer advocate at Google Cloud Platform, who says too many IT leaders are debating how to manage IT operations and workflows, when their businesses are being hit with unprecedented disruption. "CI/CD is a done deal -- like 10 years ago it was a done deal," he said in a recent podcast with CTO Advisor's Keith Townsend. "There is nothing to figure out in that domain. A lot of people talk about DevOps, and there may be some culture changes, in number of people who can do it or are allowed to do it. For me, that is the table stakes. CI/CD, DevOps; we have to say, listen, figure it out, or go work with another team outside this company to figure it out."

116 comments

  1. heads were removed from anuses by Anonymous Coward · · Score: 5, Insightful

    So somewhere along the way people figured out again that quality of software is more important than the speed in which new features are pushed out the door.

    I guess the cranio-rectal inversion over devops crap is finally coming to an end.

    Next will be when everyone moves their stuff to an "internal" cloud. Just like when people moved off of timeshare mainframes to computers on premise.

    1. Re:heads were removed from anuses by omnichad · · Score: 2, Funny

      Next will be when everyone moves their stuff to an "internal" cloud. Just like when people moved off of timeshare mainframes to computers on premise.

      Nah - caching/boost server on premises. So you can have a thin cloud server to go with your thin clients.

    2. Re:heads were removed from anuses by xxxJonBoyxxx · · Score: 1

      >> next will be when everyone moves their stuff to an "internal" cloud

      You giggle but this is actually a theme in corporate America. Typically it's the legacy VMware team that rebrands itself as the internal cloud, but I've also seen an attempt or two to build out a new virtualized stack on top of Xen or the like. It's even kind of real for teams that virtualize at the container level.

      I'm just happy to not have to pay for this stuff myself. It seems most companies are currently losing their shirt on duplicated systems that persist for years during "cloud migrations", even though they just migrated everything (and dealt with the duplicate systems thing) to get 80% of their stuff on VMware.

      However, life as someone who helps over-budgeted IT departments move toward the next shiny object can be quite good.

    3. Re:heads were removed from anuses by Dayze!Confused · · Score: 4, Insightful

      Nope, turns out this person is arguing for software quality. They think CI/CD is settled in the sense that it isn't going away, stop trying to change it, move on to other things like the latest buzz words. I believe CI/CD is what has made software shit these days. Companies start by promising x number of features, and rush it out the door after x-y>0 features are partially done, then CI/CD with users as their beta testers. You no longer have a finished version and solid piece of software, and no way to stick with a particular version, which is especially true on the walled garden of iOS. Gone are the days of a solid product coming out that stands on its own for years without the need for anything but occasional security updates.

      --
      "All tyranny needs to gain a foothold is for people of good conscience to remain silent." [Thomas Jefferson]
    4. Re:heads were removed from anuses by slack_justyb · · Score: 1

      Just like when people moved off of timeshare mainframes to computers on premise.

      Good God, have you not been in this industry long? This kind of tick tock is literally the name of the game. Be ready in sixty years when everyone touts whatever they call cloud in 2080.

    5. Re:heads were removed from anuses by Anonymous Coward · · Score: 2, Insightful

      It was a thing, pre-internet.

    6. Re:heads were removed from anuses by pdxtabs · · Score: 1

      You can say "we practice continuous deployment" without saying "to production servers." CI/CD is about your development branch never getting so shitty that the average engineer can't do meaningful work on it. Or, the other thing that used to happen: make a bunch of components and then try to stitch them together at the end; act surprised when nothing works.

    7. Re:heads were removed from anuses by Anonymous Coward · · Score: 0

      It seems most companies are currently losing their shirt on duplicated systems that persist for years during "cloud migrations"

      The number of contracts I've worked where I immediately saved thousands of dollars per month by actually auditing the fucking clowwwwwd accounts that "aren't used anymore" is ridiculous.

      I understand the trap - "Hey, let's keep this around until we're sure this new shiny thing works!"

      It's a stupid trap. Stop that.

    8. Re:heads were removed from anuses by Anonymous Coward · · Score: 0

      A shop that says 'done deal'. That may be true for Google.

      For most shops CICD is not even a glimmer in their eyes. They do not do it at all. They are busy fixing all the shit that CICD could help them catch up front.

      Devops can be done very well or very badly like all processes. For example. Waterfall works. *IF* you stick to the process and do not drag it along like a dead horse that you think will someday walk again. The idea is to pick a process that works well for you and can be improved upon.

      Many times 'devops' is just a way to knock over the artificial barrier between developers and customers. When you have devs building shit no one needs and fixing bugs that they find interesting you end up with 'broken' software. Most of the time most processes are about get out of your hole and talk to your customers find out what they really care about. CICD lets you try things out quick and fail quick instead of dragging dead weight along that none of your customers care about anymore. They may have cared 2 years ago. Now they have grown tired of waiting and have moved on either with their own process to work around you or the business opportunity is long gone.

      My goal in devops is not to get called at fucking 3AM. If I can pull that off the software actually works and does what it is supposed to do. The CICD devops process enables that. Mostly the whole devops thing says 'yep that screwed up and died lets not do that again'.

    9. Re:heads were removed from anuses by 0100010001010011 · · Score: 1

      I work in automotive embedded and I can't wait to get our CI server up and running. I think a lot of people forget what the 'good ole days' were like.

      We'll have people working on their own branches and then on Friday merges we have the whole thing fall apart. Meaning people don't have binaries to test until at least Saturday. Then the whole process starts over. ClearCase being a cluster doesn't help.

    10. Re:heads were removed from anuses by Anonymous Coward · · Score: 0

      If they're shipping the product without solid stable features that make a usable product then no "process" or "style" of management will change that. That's just a bad manager. They will purposely sabotage any process to get it out the door earlier.

    11. Re:heads were removed from anuses by angel'o'sphere · · Score: 0

      CI means continuous integration, not improvement.
      It is automated build and automated test of a piece of software as soon as the source code repository changes.

      In other words: it is an QA measure,

      CD means that software is either promoted in form of binaries into a staging repository or deployment on test servers.

      A better way to ensure software quality we don't have right now.

      If your CI/CD does not deliver perfect quality software, then either your developers suck and write bad tests, or the integration tests don't work, or you have bad specifications for the development and testing.

      You no longer have a finished version and solid piece of software, and no way to stick with a particular version, which is especially true on the walled garden of iOS
      That is utter nonsense ...

      --
      Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
    12. Re:heads were removed from anuses by Chris+Mattern · · Score: 1

      It was a thing, pre-internet.

      It was? I started out in an old-fashioned IBM S/370 shop, and I remember the steady stream of PTFs and PUTs the system programmers would get...

    13. Re: heads were removed from anuses by Anonymous Coward · · Score: 0

      A steady stream wasn't nightly.

    14. Re:heads were removed from anuses by Junta · · Score: 2

      Of course the change now is that the standard was to have PTF type updates that would fix problems without changing functionality.

      Now developers aren't forced to do that because they are too cool for that, and roll out fixes alongside functional changes that also aren't baked yet and bring new problems, so you don't get to stop and say "hey, you *almost* got it right here, I'll just take fixes from this version, thanks"

      In this era of using software remotely, you can't even choose to ignore updates and stick with an abandoned version, it's just gone.

      Of course there are enterprise vendors still doing this, but developers will bitch about being asked to support that 'ancient' platform that first released 3 years ago and screw it up by ignoring all the content and replacing with the random untested assortment of latest crap from gems/pypi/npm/whatever wtogether with a random build of the language runtime that partially screws up the system paths.

      It's just not 'cool' to have maintenance branches, so.. here we are.

      --
      XML is like violence. If it doesn't solve the problem, use more.
    15. Re: heads were removed from anuses by Anonymous Coward · · Score: 0

      "Perfect quality software."

      Hahahahahahahahahahaaaaaaaaahahahahahahaha!

      You can ensure your tests cover 100% of your code and your assertions could still be pointless or absent entirely. Or miss the point of what you're trying to do.

      Or your code could be Enterprise Spaghetti. Or the external system you integrate to could be unpredictable garbage. Or it could have an awful user experience. Or it could be completely unreliable when the network flakes in a way your tests didn't anticipate.

      Trying for perfect quality software is a fool's errand. Just try to keep it easy to fix when it breaks. Because it will absolutely break.

    16. Re:heads were removed from anuses by Junta · · Score: 2

      CI isn't such a bad thing, but CD tends to mean 'to production' for a lot of folks.

      Also, while CI isn't in and of itself a bad thing, it does encourage a lot of organizations to skimp on vluable QA, made overconfident by misleading phrases like '100% test coverage' and deciding that means they automated the need for QA.

      Also, by the same mistaken thought, they don't need maintenance branches anymore because the first release was necessarily bug free because those unit tests passed, so no need to think about whether a change makes sense to work into an older functional branch.

      --
      XML is like violence. If it doesn't solve the problem, use more.
    17. Re:heads were removed from anuses by Junta · · Score: 2

      Of course with CI, you can still have this phenomon. This is about midnset and developer strategy, You can still have developers sitting on their private branches and at the last minute end of conflicting with each other. Of course a sane process says they get to sit that week out, but in all likelihodd, one of those has some change that simply "can't wait" even a week, so the whole train gets held up despite the sane thing of punting that developer work for a week,,

      --
      XML is like violence. If it doesn't solve the problem, use more.
    18. Re:heads were removed from anuses by Junta · · Score: 1

      In general, the reality is that the biggest losers are the ones *obsessed* with appearing to conform to .

      Recently the great corporate overlords came to make my team "go Agile" to improve our productivity. Of course we already were 'Agile', but the company brought in 'certified Agile' consultants to retrain project managers, and frankly for somewhat plausible cause since there are some really crappy teams that no one wants to admit just aren't qualified and would rather a process magically fix those teams.Prior to the mess my team had continuous integration with unit tests prior to our QA team, our QA team had access to builds as changes were made, the average time from a bug opened to a test build available for QA team (after CI and all) was about 2.5 hours. Our release cycle dictated shippable increments and punting functions not ready.

      What did the 'Agile' project management team and consultants do? Throttled us to weekly builds (wtf??), demanded 100% commitment on functionality 1 year in advance, and started saying functions *must* go in and we need to slip dates if we didn't get a feature completed in the timeframe we thought (despite no external stakeholder expecting the features), and imposed that no customer request could be satisfied sooner than about 9 months from asking, even if minor. They started 'helping' by preventing developers from talking to customers and talking to them instead. The end result is a bizarre situation where the QA team has asked dev team for a secret build server to provide fixes like we used to without telling management, and customers no longer going through official channels because they no longer reach developers, but sending personal emails to the dev email addresses they remember, begging for help and betas. That latter thing of course produced an introducing graph, a spike in customer support traffic and then a decline as informal channels triumphed over official, so their take away is 'greater customer sat' (despite declining sales because 'market pressures')

      But hey, now we are more 'Agile' because we use trendy words like 'user stories', 'scrum', and 'sprints'. Long story short, crappy teams and crappy project management will be crappy, but they dream of easy fix through dumping money on consultants without fixing the fundamental problems.

      --
      XML is like violence. If it doesn't solve the problem, use more.
    19. Re:heads were removed from anuses by 0100010001010011 · · Score: 2

      Absolutely. But the "CI" part helps to automate it. Our builds are in the ~50 minute range. Individual developers build on their own machines. There's no 'safe' build machine. At least with a CI setup they get immediate "You broke it dumbass" feedback.

    20. Re:heads were removed from anuses by datavirtue · · Score: 1

      People have been screaming about DevOps because hardly anyone was doing it right and "not doing it right" led to a lot of bullshit in people's lives and generally made IT look bad. Unfortunately this is still the case...but the post is correct in that the focus is on improving these processes to the detriment of the business at this point. If you haven't got to some sort of DevOps sanity by now you are probably never going to. ....and yeah....the "internal" cloud is the next big thing. The same people who can't get their DevOps shit together are the same ones who can't exploit cloud services and are most likely going to deploy on to a IaaS abstraction layer that will let them move to the cloud when and if they need to without modifying their code base or configurations. Pivotal Cloud Foundry for instance.

      --
      I object to power without constructive purpose. --Spock
    21. Re:heads were removed from anuses by Billly+Gates · · Score: 1

      Nope, turns out this person is arguing for software quality. They think CI/CD is settled in the sense that it isn't going away, stop trying to change it, move on to other things like the latest buzz words. I believe CI/CD is what has made software shit these days. Companies start by promising x number of features, and rush it out the door after x-y>0 features are partially done, then CI/CD with users as their beta testers. You no longer have a finished version and solid piece of software, and no way to stick with a particular version, which is especially true on the walled garden of iOS. Gone are the days of a solid product coming out that stands on its own for years without the need for anything but occasional security updates.

      Funny I thought CIO's were about labeling IT as a cost center with a focus on saving. Or better yet being eliminated by having ther HR or finance department run IT who get bonuses by outsourcing to the cheapest country as possible and using 10 year old software and never updating as that is out of the contract with TATA.

      Call my cynical and perhaps this has changed from last decade and the last recession but that is how I remember it.

    22. Re:heads were removed from anuses by Billly+Gates · · Score: 1

      What the hell is integration? It is so 1990s buzzword but I do not know what it means. It sounds like an excute to only buy from Microsoft or something as it works together if I recall PHBs using integration constantly.

      QA is going away thanks to Agile development. That is why Microsoft has no QA at all whatsoever with Windows 10. With Agile/Scrum the developers and users do the QA with happy faces/sad etc.

    23. Re:heads were removed from anuses by datavirtue · · Score: 2

      Show me someone who has "100% test coverage" and I will walk over to the closet and let the unicorn out.

      --
      I object to power without constructive purpose. --Spock
    24. Re: heads were removed from anuses by datavirtue · · Score: 1

      " Or it could be completely unreliable when the network flakes in a way your tests didn't anticipate."

      This. Nothing boils my blood quicker than a developer blaming the infrastructure for a bug in their software. I snap back: "Why did you trust the network? Never trust the network or the database!" "I know, I know." "Then why are you back here? Go fix your code!!"

      --
      I object to power without constructive purpose. --Spock
    25. Re:heads were removed from anuses by EvilAlphonso · · Score: 1

      Similar experience with big iron... I saved my previous gig $2MM/year by decommissioning a farm of legacy SF15K and SF25K that hadn't seen any action (other than patching) for 6 years. Nobody actually had the guts to pull the plugs, so the company was still paying support for them.

    26. Re:heads were removed from anuses by johannesg · · Score: 1

      Next will be when everyone moves their stuff to an "internal" cloud. Just like when people moved off of timeshare mainframes to computers on premise.

      Actually this already happened. A few weeks ago our salesman happily reported that a customer would be storing data in their 'private cloud'. I've seen that private cloud; it happens to be on-site at the customer premises. It looked a lot like a server rack to me...

    27. Re: heads were removed from anuses by Anonymous Coward · · Score: 0

      FFS you really are a part of the fucking problem.

      It is possible to write software that is easy to fox, reliable, and isn't sensitive to "network flaking in a way..." because it has meaningful application-level state that is checked on point to point communication, and so forth. It might fail, but not routinely and not by reasonably foreseeable events. So overall the risk is low.

      And that is work worth doing.

    28. Re:heads were removed from anuses by turbidostato · · Score: 2

      "CI isn't such a bad thing, but CD tends to mean 'to production' for a lot of folks."

      That's because it's either "to production" or it is not CD at all.

      Of course, "to production" may mean different things to different people. For a VAR/COTS mill it means, "that's what we are going to sell to our next customer", for an internal team, it will mean "deployed and tested to our production servers" but, still, "delivery" means "delivery"; if it's not delivered, then it can't be D, continuous or otherwise.

      "Also, while CI isn't in and of itself a bad thing, it does encourage a lot of organizations to skimp on vluable QA"

      Bad managers are always bad managers. Where in "continuous" is hidden the concept "let's take critical steps out of our flow"? Hint: nowhere.

      For the ones in the know it has been always the same: let's think out a good process matching our goals, let's test it to be valid, let's take out people from the process where people adds no value, rinse and repeat. That's "continuous whatever". Now: start with a bad process and all that "continuous" can add is "fucking it faster" -which, provided you have good brains at the rudder, could be a good thing on itself except, of course, you have bad processes in place because you have bad brains at the rudder to start with.

      And then, there is the "devops" thingie. Look at the "agile manifesto" and all the other "foundational" concepts on these fields. All of them go to the same thing: "let's understand this is still not a "proper" engineering -and maybe never will be, but more a craftmanship, so let's understand this is more about people and empowering people than it is about processes -just the opposite to taylorism". But then, this is not a "silver bullet", because there are no "silver bullets" and where the risk of taylorism is siloism and the unability of letting your most valuable craftsmen to provide you the most value (and, in the most extreme cases, letting your people dealing with procedures and paperwork so much as the output of real value being reduced to zero), the risk of "agilism" as empowering people, is empowering the wrong people -which given the lame state of current IT, both at the technical and managerial level will be "most of them" and it may be the case that, given Sturgeon's law, there's no way to avoid that at a scale.

      "Also, by the same mistaken thought, they don't need maintenance branches anymore"

      Again, it depends: a VAR/COTS requires maintenace branches, an internal product/service probably does not and a SaaS offer may or may not, depending on those pesky details.

    29. Re:heads were removed from anuses by turbidostato · · Score: 1

      "I've seen that private cloud; it happens to be on-site at the customer premises. It looked a lot like a server rack to me..."

      Do you think a public cloud looks much different than a server rack, only on other's datacenter?

    30. Re:heads were removed from anuses by zifn4b · · Score: 1

      So somewhere along the way people figured out again that quality of software is more important than the speed in which new features are pushed out the door.

      The company I work for seems to be in its death throes with a person leaving the company every week. They have just now started to get a clue that low software quality has a relationship to losing your customer base and escalated support issues and consequently your reputation as a software vendor. The thing that I see neglected the most in software companies is subject matter domain knowledge. If you don't know what your software needs to do, you can't communicate that to the people that build it and that means that what you build will not be what the intended customer base wants. Ultimately, that means FAIL. It's a garbage in/garbage out problem. It's not even a technical problem for the most part except scalability.

      Ultimately, the real root cause of this problem is hiring on the cheap instead of paying for real talent. I wonder when companies will start getting a clue that you can't find unicorns in the bargain bin.

      --
      We'll make great pets
    31. Re:heads were removed from anuses by Anonymous Coward · · Score: 0

      So somewhere along the way people figured out again that quality of software is more important than the speed in which new features are pushed out the door.

      It isn't just about "new features":

      Kaizen (), is the Japanese word for "continual improvement". In business, kaizen refers to activities that continuously improve all functions and involve all employees from the CEO to the assembly line workers.

      * https://en.wikipedia.org/wiki/Kaizen
      * https://en.wikipedia.org/wiki/Toyota_Production_System#Continuous_improvement

    32. Re:heads were removed from anuses by Anonymous Coward · · Score: 0

      Nope, turns out this person is arguing for software quality. They think CI/CD is settled in the sense that it isn't going away, stop trying to change it, move on to other things like the latest buzz words. I believe CI/CD is what has made software shit these days.

      If that is true, then people are forgetting what the "I" in "CI" stands for:

      Kaizen (), is the Japanese word for "continual improvement". In business, kaizen refers to activities that continuously improve all functions and involve all employees from the CEO to the assembly line workers.

      * https://en.wikipedia.org/wiki/Kaizen
      * https://en.wikipedia.org/wiki/Toyota_Production_System#Continuous_improvement

      If you can only push to production once a year or once a quarter, then you can only fix bugs once a year or once a quarter. If you can only release code once a year, then there's more pressure to put a non-fully baked feature in because you'll have to wait for a long time before you get the next chance.

      If you're releasing every month, fortnight, or even every week, then missing a release for a particular feature isn't as big of deal because you'll get to on the next go-around.

      The problem you're describing isn't a problem of process, it's a problem of not caring about quality and suffering feature-itis. Releasing shit code happened before CI/CD became popular, but then it was much harder to fix because the next opportunity for new code happened so infrequently.

    33. Re:heads were removed from anuses by angel'o'sphere · · Score: 1

      QA is going away thanks to Agile development
      This is kind of an oxymoron. QA, especially automated testing is the corner store of agile methods.

      With Agile/Scrum the developers and users do the QA with happy faces/sad etc.
      What is that supposed to mean?

      --
      Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
    34. Re:heads were removed from anuses by angel'o'sphere · · Score: 1

      100% test coverage is easy, with unit tests. Which are close to meaningless.
      Interesting is 100% code coverage with acceptance tests/ui driven tests/integration tests.

      --
      Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
    35. Re:heads were removed from anuses by Junta · · Score: 2

      The curse of all of this is that good teams have always naturally done things that lead to good product, and then some folks comes along to try to 'distill' out the magic of those successes into useless aphorisms and make a consultancy industry out of it. Complicated by ditching the unprofitable pieces (for example the agile manifesto among other things stressed communication rather than processes and tools, but an Agile consultant will ditch that and focus on tools and processes because they can make money off of tools and processes are what sells to the management).

      So 'continuous blah' and 'devops' and 'agile' are all afflicted by the curse of being the consultancy of the month and as such are applied to bad teams and leave bad tastes in the mouths of folks. Also, the business is held up by taking the no true scotsman both ways. A project that did the Agile or DevOps or CI/CD thing and failed, well they didn't do it right, so it wasn't "truly" whatever it was. By the same token, a project succeeds, it is deemed to have done it right, even as it makes no sense in the context of the purported philosophy..

      --
      XML is like violence. If it doesn't solve the problem, use more.
    36. Re:heads were removed from anuses by Junta · · Score: 2

      I guess I've been blessed in that I've never worked in a team that wasn't naturally continuously building and testing their changes. I couldn't imagine being more than a few minutes of being able to see the result of my work. So for me, CI comes into novel play only when merge requests happen that start exposing subtle conflicts between privately worked branches of work.

      I suppose I could imagine someone spending a long time without even building and testing, but on the other hand I can't imagine that willingness translating to a good developer even if that particular bad habit is automated away.

      --
      XML is like violence. If it doesn't solve the problem, use more.
    37. Re:heads were removed from anuses by ketomax · · Score: 1

      I think he means the feedback system with emoticons that some beta Microsoft products sported.

    38. Re:heads were removed from anuses by AuMatar · · Score: 1

      Still meaningless.

      if(A) {
        foo()
      }
      if(B) {
        bar()
      }

      You can have 100% without testing the combo of foo and bar. WHich could break everything. Code coverage numbers just give a false sense of security.

      --
      I still have more fans than freaks. WTF is wrong with you people?
    39. Re: heads were removed from anuses by Anonymous Coward · · Score: 0

      Please name the "Agile" consultancy involved. They deserve public mockery, and I'd like to know who to avoid.

    40. Re:heads were removed from anuses by Anonymous Coward · · Score: 0

      If your CI/CD does not deliver perfect quality software, then either your developers suck and write bad tests, or the integration tests don't work, or you have bad specifications for the development and testing.

      Says someone who has never developed software in complex systems level software in a highly concurrent system with performance requirements (esp. requirements which place tight restraints on performance regressions for customers who do not use new enhancements as such requirements can sometimes require compromising design ideals).

      In such systems, very obscure race conditions involving the interaction of many different subsystems can (and do) exist. The fact one of these is missed during the design phase of a feature whose implementation is spread across several subsystems does not mean the "developers suck" -- it means they are human. In such complex systems, the design of some major features requires the coordination of many developers across many subsystems and no one person can possibly know each subsystem well enough to understand every detail of each.

      If the designers miss such a condition, no unit test in the world will reliably identify it (unit tests test that a unit works as intended -- in these cases, every unit can work exactly as "intended" but the units just don't fit together in some combinations of timing of events). Similarly, focused integration tests (which test reasonable combinations of functions and exception events) won't catch it. Obviously all combinations can't be tested because simply writing a set of tests that did that would take the entire staff of the largest company in the world between now and the heat death of the universe.

      Following code freeze, such systems need to be subjected to stress testing, often for months, on a lot of hardware to build confidence and to expose the more obvious race conditions. For example, those race conditions that occur on the average, under stress w/induced exceptions, once every 1000 years on a single node (you hope that your customers running your software on many 1000's of nodes aren't encountering as many exception cases in their production so they never actually encounter such bugs - else your phone will ring at three in the morning when your customer's CIO has called your CEO and demanded "the best people" be working on the bug). Of course, these heisenbugs which also occur very rarely often take man-months or even man-years of developer time to isolate and resolve once it's noticed that they were encountered.

      One simply can't run such stress tests continuously because long before they finished the results become obsolete - nor would it make a lot of sense as the "after code freeze" test is still needed.

      But, yes, for a simple small system (perhaps a few hundred thousand lines of single threaded code) it may be possible for unit tests and integration tests to result in a seemingly "perfect quality" system if it's built by excellent developers. However, no company is going to pay enough to attract excellent developers to build such a system and few if any excellent developers would be interested in working in such a system. There simply are not enough excellent developers available to waste on such systems.

      Of course, the only way that a system can be "perfect quality" in the real world is because there was a LOT of luck involved -- I don't know of a single perfect complex system that was ever built. ("perfect quality" would be no bugs in requirements through implementation exist).

    41. Re:heads were removed from anuses by angel'o'sphere · · Score: 1

      Depends what you conclude from them :)
      100% coverage is certainly better than 50%.

      Coverage is important for future refactorings, less for the quality of your actual tests. I'm not a big fan of code reviews, but tests I always review or insist that they get reviewed.
      For Java there are byte code moephing tools that that exchange all test operations by its counterpart. If your tests are 'somewhat good' all should fail after morphing.

      --
      Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
    42. Re:heads were removed from anuses by Anonymous Coward · · Score: 0

      Are you sure that some of those incompetent teams aren't there through nepotism? Maybe the consultants were really told to provide cover fire for the CEO's nephew or niece. All those other insane changes sound perfectly rational if the goal is to protect someone's feelings by preventing people you from out-performing.

    43. Re:heads were removed from anuses by datavirtue · · Score: 1

      Exactly. Developers get the business to let them waste time on writing unit tests because they hear "testing" which sounds really good and like it may be something that could prevent the constant bumbling that follows each release. I sit there mumbling something about "waste of time" under my breath and ask how they are going to ensure that this test suite is going to be maintained and that it will actually test anything we care about. The developers admit to the pitfalls, the air lets out of their sails, and they go ahead with it anyway because it is what every good developer is supposed to do. To this day, after years, there is no test coverage.

      Pretty much the same with QA. They test what the developers tell them to test and seek guidance from them on how and what to do. Surprise, surprise, we discover how the application really works when it hits production.

      --
      I object to power without constructive purpose. --Spock
    44. Re:heads were removed from anuses by AuMatar · · Score: 1

      No it isn't. 100% means every line is run once. That 50% case may only run part of the code base, but it may cover a subsystem completely with all combo of options. I'll take the second over the first. Which is why code coverage numbers are worthless. We have no idea which of those two situations it is based on the numbers.

      --
      I still have more fans than freaks. WTF is wrong with you people?
    45. Re:heads were removed from anuses by johannesg · · Score: 1

      No, but why would you call a server rack a "private cloud"? What weird, pretentious nonsense is that?

    46. Re:heads were removed from anuses by turbidostato · · Score: 1

      "The curse of all of this is that good teams have always naturally done things that lead to good product, and then some folks comes along"

      Not "some folks", but "some folks with authority", aka "management". Of course, everybody can impact a project/service/product/whatever but, of course too, the higher the individual in the corporate ladder, the higher his impact on the results. And it's my opinion that if the average level ot IT people is lame, the higher you climb the corporate ladder the more unfit for the position the people tend to be (I could write quite a lot on why this is the case, but it would be to long and I don't feel in the mood).

      "So 'continuous blah' and 'devops' and 'agile' are all afflicted by the curse of being the consultancy of the month"

      There would be no "consultancy of the month" if the one paying for such consultancy knew things better (so see above).

    47. Re:heads were removed from anuses by turbidostato · · Score: 1

      "No, but why would you call a server rack a "private cloud"? What weird, pretentious nonsense is that?"

      The pretentiousness, if any, is not in the "private" part, but in the "cloud" one.

      I've been in the "cloud thingie" basically from the beginning and I've seen how the term has derived from something quite significant to the vague term it quite quickly became but then, if it allows for self servicing on demand of IT resources (mainly infrastructure, but everything above can also work), then I certainly call it "cloud", private or otherwise so, yes, a, say, internal openstack deployment, sensibly configured and used, I would call an "internal cloud" even if in the end is not much more than an on-premises rack (or a bunch of them).

    48. Re:heads were removed from anuses by angel'o'sphere · · Score: 1

      I don't understand what you mean with your 100%
      How can a line of code be run only once without covering all options that lead to running it?

      --
      Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
    49. Re:heads were removed from anuses by AuMatar · · Score: 1

      Because the 100% measures lines independently. It doesn't mean that combinations of branch statements/loop iterations are tested. Without which, you claim to be testing 100% but are really testing some unknown but much lower percentage of actual code paths. Just running each line once has 0 actual value, it doesn't mean the code is well tested. I would rather have a suite that actually tests some subset of the code exhaustively than a suite that runs each line once and claims that everything is perfectly tested.

      Actually you're perfectly illustrating my point of why the number is garbage. You actually think 100% code coverage means you have good tests. Its completely tangential.

      --
      I still have more fans than freaks. WTF is wrong with you people?
    50. Re:heads were removed from anuses by angel'o'sphere · · Score: 1

      You make no sense, sorry.

      If you have 100% code coverage, you have 100% coverage of all "if's" "else's" and "loops", too.

      Or how do you think I magically test a line in the middle of an if block without testing the line with the "if"?

      I would rather have a suite that actually tests some subset of the code exhaustively than a suite that runs each line once and claims that everything is perfectly tested.
      What is the point of testing the same line of code multiple times with different tests? Unless: it is an "if" or the "loop condition". Which you obviously test with the corner cases.

      The only thing you are right with: if I test an If only for success and not for failure, I have 100% coverage but miss an important case. Similar for loop conditions.

      --
      Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
    51. Re:heads were removed from anuses by AuMatar · · Score: 1

      You're either not trying, or you're a really bad programmer. Let's try this:

      if(x) {
      //enable Feature A
      }
      if(y) {
      //enable Feature B
      }

      What if Feature A and B are incompatible. Turning on B and A at the same time breaks the entire app. The fact we can try to enable both at the same time is a bug.

      You can have 100% code coverage without ever testing what happens if x and y are both true, by testing x=true y=false and x=false y=true. You would then test each line, and say you had 100% coverage. But you wouldn't be testing every way the app can be run, and would be missing bugs. The 100% number is pointless, it isn't well tested.

      If you still don't get it- find another profession, you're not competent in this one.

      --
      I still have more fans than freaks. WTF is wrong with you people?
    52. Re:heads were removed from anuses by angel'o'sphere · · Score: 1

      I would say, you are super bad in explaining.

      No, I don't get it.

      The above code you test with
      x,y = false, false
      x,y = true, false
      x,y = false, true
      x,y = false, false

      What is wrong with that?

      --
      Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
  2. wtf by Anonymous Coward · · Score: 0, Funny

    I have chickadees at my birdfeeder but instead of making normal chickadee sounds they go "chick-a-DONK-DONK-DONK". And a lot of the time if you're coming or going they fly at you and attack you with their little beaks like miniature kamikaze pilots. And if they miss you they circle back and shit on you. I'm starting to think the little f**kers have bird flu or something, I've never seen chickadees out for blood like this. Hell, they're usually friendly little guys. I think I'm going to get some of those sticky glue mouse traps and mount them on the bird feeder, see if I can catch some of those bastards. I wonder if they could have cross-bred with killer bees and made some kind of 'Africanized' species? Maybe I should try to catch a specimen.

  3. Uh Huh by Anonymous Coward · · Score: 1

    I have yet to see to a CI/CD shop actually implement this in such a way that people can safely commit away.

    I'm sure life is easier when you don't actually have customers that care about state or you just shit on your customers continuously. Google wins on both counts here!

    1. Re: Uh Huh by Anonymous Coward · · Score: 0

      I work in the financial industry. Every day I wish we sold clickbait like everybody else so we could be part of the 24/7 party that is continuous deployment. But our customers need our shit to actually work, not just make it past the testing robots.

    2. Re: Uh Huh by Anonymous Coward · · Score: 0

      Agreed. Iâ(TM)m so tired of all these companies that only do one or two things setting the standard for how the rest of us operate. If all I did was stream video to customers and bill them monthly my job would be gravy.

    3. Re:Uh Huh by Anonymous Coward · · Score: 0

      Well, we split our whole environment into a pipeline of dev1/dev2/dev3/staging/prod systems and do CI/CD for multi-product systems development. But we still have more conventional QA/CM processes to approve the subset of configurations that go to production.

      Obviously the first layer of dev is the actual developer workstation environments where they play in their own private sandbox. They can hack and run local tests even before they commit something. The second layer is that product repos have a transient VM-based test suite run on every commit/pull request/merge. This includes a trimmed down full stack integration appropriate to that product's test suite, e.g. for a web front-end product, we spin up a temporary DB, services, and front-end static files in a temporary web service VM, then run test suites that provision test data into the service stack and enslave a browser to run the front-end apps through full user workflows. These suites give us red/green status for each repo, branch, and tag.

      The third dev layer is standing integration systems. They are provisioned once in a while with the same automation processes as a production system, but hourly updatet to master/head revisions of every constituent product. These are multi-user deployments visible across the org. The VMs are not provisioned hourly, but persist for weeks with the same evolving test databases. Only the stateless software stack is updated hourly. This is where dev, QA, project managers, and even friendly users/stakeholders can interact with the latest stuff for internal demos or problem determination.

      The staging layer is like production, but it's for a configuration engineer to run dress rehearsals of a fully tagged release candidate. Our tags are for latest stable revisions passing the integration tests done here in staging. As another nod to agile/CI, we really only fix things by moving forward. The engineer can only choose to hold back at a previous stable tag or move forward to the next one that incorporates all fixes (and new features) that they are ready to live with.

      The prod layer is obviously the real stuff in front of production users. It's where successful release candidates from staging get repeated for real. This hybrid approach works well, in my opinion.

      But, when the higher up management people are unhappy with the delays in a pet feature going into production, they often concoct imaginary explanations. For example, they will assume that it must be some overhead in our deployment tools that makes it "too hard" to just release the latest. They cannot comprehend that it is actually their own staff holding back the release, because they have deemed it too high risk this week. If your management gets too envious of those clickbait vendors who can really push broken code to production with impunity, they may subconsciously try to steer the organization in that direction, while blinding themselves to the fact that this is a strategic change in risk mitigation/posture.

      It can get very confrontational if managers try to override their delegates' decisions and force a risky update, without that same manager clearly accepting the responsibility for the decision. Nobody wants to live under the pressure of "move fast and break things" if they feel that their performance review also requires "don't break things."

  4. But the testing goat said... by Anonymous Coward · · Score: 1

    I just started reading "Test-Driven Development with Python: Obey the Testing Goat: Using Django, Selenium, and JavaScript" by Harry J.W. Percival. Now CI/CD is obsolete.

  5. The future is NoOps by Anonymous Coward · · Score: 0, Insightful

    With Amazon Lambda and other microservices, you just need HR to hand out IAM accounts, and a company really doesn't need an IT staff whatsoever. Just some CI/CD mechanism to get pushes in production, and that is basically it.

    Ops is dead. Who needs to rack and stack physical servers when the cloud takes care of that, and far cheaper. Who needs OS guys, app guys, net admins, and DBAs when serverless services replace all this?

    Lets be real... the future is NoOps. Pay your Amazon bill, and they take care of your IT infrastructure.

    1. Re:The future is NoOps by sexconker · · Score: 5, Insightful

      With Amazon Lambda and other microservices, you just need HR to hand out IAM accounts, and a company really doesn't need an IT staff whatsoever. Just some CI/CD mechanism to get pushes in production, and that is basically it.

      Ops is dead. Who needs to rack and stack physical servers when the cloud takes care of that, and far cheaper. Who needs OS guys, app guys, net admins, and DBAs when serverless services replace all this?

      Lets be real... the future is NoOps. Pay your Amazon bill, and they take care of your IT infrastructure.

      The cloud is NOT cheaper. Amazon is expensive. And you never know how much you're going to pay.
      Even hosting a simple static website is a nightmare. They have about different 40 products for simple domain name management alone.

    2. Re:The future is NoOps by Anonymous Coward · · Score: 1

      No future in ops means nobody studies it. Who does Amazon hire in 30 years to keep it all going?

    3. Re:The future is NoOps by phantomfive · · Score: 1

      Even hosting a simple static website is a nightmare.

      Put it on S3, you're done. Use whatever registrar you want. If you need help figuring out how to point your DNS to S3, then there are forums and stuff. Ask on Stackoverflow.

      --
      "First they came for the slanderers and i said nothing."
    4. Re:The future is NoOps by Anonymous Coward · · Score: 1

      How does one get online to the cloud without a network, I wonder?

    5. Re:The future is NoOps by sexconker · · Score: 3, Informative

      Ha!

      Here's the "simple calculator" that doesn't even cover all of the services:
      http://calculator.s3.amazonaws...

      Put it on S3? S3 is storage! You need it to be on EC2. Possibly behind Beanstalk. Oh, you want to actually make use of the fancy cloud features for internet-accessible shit? You'll need Route 53, too, and the Elastic Load Balancer.

      https://aws.amazon.com/s3/pric...
      https://aws.amazon.com/ec2/pri...
      https://aws.amazon.com/route53...
      https://aws.amazon.com/elastic...

      Beanstalk is free, though!

      Take a look at this fucking list. https://i.imgur.com/nBasljK.pn...

    6. Re:The future is NoOps by phantomfive · · Score: 1

      If you want to host a static website, all you need is S3. That is it. Don't forget to make your files public-readable. EC2 and load balancers are for dynamic content.

      For the most part, you shouldn't be using the fancy cloud features. Keep it simple. The only way to have a reasonable system is to keep it utterly as simple as possible. Anything that ties you into Amazon proprietary products should be ignored, although Amazon will try to convince you to pay for them. Anything too fancy will come back and bite you.

      --
      "First they came for the slanderers and i said nothing."
    7. Re:The future is NoOps by Anonymous Coward · · Score: 0

      You were under the impression that phantomfive understands IT.

      Today enlightenment found you.

    8. Re:The future is NoOps by Anonymous Coward · · Score: 0

      I officially declare: you sir, know absolutely nothing about Amazon web services!

    9. Re:The future is NoOps by XXeR · · Score: 1

      The cloud is NOT cheaper. Amazon is expensive. And you never know how much you're going to pay.

      When did Slashdot become full of people that talk in absolutes like this, and then get moderated insightful? It barely warrants a response, but I don't want to say that without any context.

      Cloud is definitely not the answer for everyone, and AWS isn't always the right answer even when cloud is. It completely depends on your situation. Stop being so narrow minded and understand that your way isn't the right way for everyone in every situation.

      As for the cost of AWS in particular, they absolutely nickel and dime. But, if you understand how it works, it's actually fairly predictable.

    10. Re:The future is NoOps by Anonymous Coward · · Score: 0

      DevOps services are not static websites, nor are they simple.

      Perhaps you are confusing DevOps with Marketing?

    11. Re:The future is NoOps by Anonymous Coward · · Score: 0

      "When did Slashdot become full of people that talk in absolutes like this, and then get moderated insightful?"

      People vote up what they want to believe. There are a lot of DevOps, IT, DBAs, etc. on this site that don't want to hear that Amazon/Microsoft/Google are coming for their jobs.

      Yes, AWS is not cheap, but it's much cheaper when you start laying people off that used to do the same work and not as efficiently

    12. Re: The future is NoOps by guruevi · · Score: 1

      I can pay three to five salaries with the cost of Amazon. And even though you have Amazon, you still need those IT people to set it up and help people use it.

      $200k/y for a service that cost $50k in hardware, $1k in power and 5% of a jr. FTE - reals savings.

      --
      Custom electronics and digital signage for your business: www.evcircuits.com
    13. Re:The future is NoOps by Billly+Gates · · Score: 1

      Mathematically it doesn't make sense how Amazon can be cheaper. So they host it instead of you. But now you have to pay Amazon and all their developers, middleman, taxes, and the same costs you paid when the servers were in your MDF.

      The only logical reason why this is taking off is Wall Street is penny wise and dollar dumb and stupid as they only count revenue in quarters. So you have one fixed cost but no spikes which make the share price go up. Where if you hosted it you would see a dip in the quarterly shareprice as you would need to upgrade hardware/software.

      So you end up paying more but evenly.

      Unless you are traded on Wall Street or small and do not have an IT department yet it sounds like a bad bet. I see Amazon is becoming the next Microsoft with proprietary API's and frameworks which glue you in too which is another red flag. Unless the bet is the Amazon APIs are so amazing and good that they give you a solution you can't get with opensource?

    14. Re:The future is NoOps by Anonymous Coward · · Score: 0

      phantomfive made it abundantly clear he was quoting the "simple static website" complaint of sexconker and how S3 alone addresses it. You are the confused one, AC.

    15. Re:The future is NoOps by Anonymous Coward · · Score: 0

      Even if you externalize to the cloud, in a lot of cases the company still has a duty to make sure it isn't reliant on external party and can continue to exist without it. (I.e. make sure there are off-site backups.)
      How do you manage this with NoOps while still ensuring confidentiality and integrity of that data?

      Proving due care and due diligence are pretty important thing once your comany gets into hot water (such as legal actions).

    16. Re:The future is NoOps by sexconker · · Score: 2

      AWS doesn't pass on the economy of scale to you. They keep it, and then some, as profit. They're not a charity.

      If you're a not a small shop, it makes sense to host yourself or colo. You can use any other datacenter for disaster recover / fallback.
      It's cheaper and you have far more control.

      AWS, Azure, etc. only make economic sense in a few niche cases. One of which is being a small, but not very small, operation who is willing to bleed money during a growth phase. Another is needing highly dynamic and geographically distributed operations NOW (not having the time to negotiate with partners).

    17. Re:The future is NoOps by Anonymous Coward · · Score: 0

      How do you know the cloud services don't log IP addresses of all the users of all the servers in there? With private servers you can more confidently promise your customers their usage is not centrally logged by some global peeping tom.

      Cloud is but another way to consolidate peeping to centralized locations. I wouldn't be surprised if the cloud providers are getting gov subsidies.

    18. Re:The future is NoOps by turbidostato · · Score: 1

      "No future in ops means nobody studies it. Who does Amazon hire in 30 years to keep it all going?"

      If they are clever (and corps don't tend to be too clever), they'll train their own people. If they don't, it'll end up more or less like Asimov's Empire nuclear facilities on the Foundation's saga: some magic tricks that nobody can repair much less enhance. But hey, it they manage to rise to monopoly position, it won't matter as nobody will have the knowledge to challenge their position, either.

    19. Re:The future is NoOps by XXeR · · Score: 1

      AWS doesn't pass on the economy of scale to you. They keep it, and then some, as profit. They're not a charity.

      I never claimed anything to the contrary, were you responding to someone else?

      If you're a not a small shop, it makes sense to host yourself or colo. You can use any other datacenter for disaster recover / fallback.
      It's cheaper and you have far more control.

      You're making assumptions here, likely based on anecdotal evidence. I have plenty of my own, and I argue that infrastructure TCO (either cloud or on-prem) is far more complex than comparing an AWS bill to an amortized set of hardware and a monthly colo bill at Sungard.

      AWS, Azure, etc. only make economic sense in a few niche cases. One of which is being a small, but not very small, operation who is willing to bleed money during a growth phase. Another is needing highly dynamic and geographically distributed operations NOW (not having the time to negotiate with partners).

      Again, assumptions. It's quite likely cloud never made sense in scenarios you've encountered, and that's fantastic for you. I currently run a large AWS footprint that would nobody from the on-prem world would call "small" if it was in their colo or datacenter, and it makes complete fiscal sense to do so. I've also run large on-prem environments...and having run both cloud and on-prem multiple times, I can guarantee you that the right answer is completely situational.

    20. Re:The future is NoOps by XXeR · · Score: 1

      Hosting anything in AWS without integrating with and properly leveraging the ancillary services they offer makes little fiscal sense.

    21. Re:The future is NoOps by Anonymous Coward · · Score: 0

      It is also cheaper to cook your own food than eat out. Yet, people eat out constantly...

      Called convenience tax. ;-)

    22. Re:The future is NoOps by geekymachoman · · Score: 1

      I remember when people talked how cheap it is.. we got 2 micro instances ( one did the work, another one was hot spare in case something happens to the first one ), and the bill at the end of the month was around $200 usd. This was sometime around 2010.

      Look what you can get for that kind of money (granted, this was a bit more expensive back then, but amazon is still the same price): https://www.hetzner.com/dedica...

      Now, it would've been cheaper, yes, if my boss fired me, his system admin, and managed all by himself. But he still needed me to make sense out of all of that, and setup everything we have / need running on these instances.

      Amazon started this "oh we're cheaper if you do X, and Y" because it's their business and they wanted the money spent their way.

  6. The problem is that there is no 'Test' in DevOps by Zubinix · · Score: 2

    I've always felt that integrating and keeping up to date test automation processes as the greatest challange in the CI/CD space. As business cycles get shorter, creating and maintaining the required set of test automation processes that can give you confidence in the final production release can be an immense challange. This together with the increasing complexity of cloud based systems has made the testing challange a really hard nut to crack.

  7. So whats the new buzzword? by Anonymous Coward · · Score: 0, Offtopic

    So whats the new buzzword?

  8. Re:The problem is that there is no 'Test' in DevOp by Anonymous Coward · · Score: 0

    Sorry, but if the projects you worked on had no integration tests, that was a problem with you and your team, or possibly the whole organization. Where I work, we have both unit tests and integration tests, not only for application code, but also for the code we use for system administration/orchestration/automation, and also for the code that supports the development environment (i.e. devops tooling).

    It can certainly get difficult, but saying that DevOps has no testing is too sweeping a statement.

  9. Humble Bundle Cloud. by Anonymous Coward · · Score: 0

    Oddly enough the latest Humble Bundle (owned by IGN) is about cloud computing.

    https://www.humblebundle.com/books/cloud-computing-books

  10. MAXIMUM OVERSNARK! by Thud457 · · Score: 1
    --

    the preceding comment is my own and in no way reflects the opinion of the Joint Chiefs of Staff

    1. Re: MAXIMUM OVERSNARK! by Anonymous Coward · · Score: 0

      Unlike local servers which have never failed, ever! Ever!!

  11. a handy guide by Anonymous Coward · · Score: 0

    Traditional development cycle -Ready, aim, fire.
    Lean - Ready, fire, aim
    DevOps - Fire, fire, fire, fire

  12. Re:The problem is that there is no 'Test' in DevOp by Zubinix · · Score: 1

    Unit and integration level tests are not enough by themselves as the gap between the systems these tests run on and the actually production system is too great, and with the complexity cloud deployments allow this gap is growing. Yes, you can run a set of smoke tests in production but that doesn't hide the fact that most of the testing is not performed in a realistic environment.

  13. Re: The future is NoOps -insightful? Lololol by Anonymous Coward · · Score: 1

    Someone has been reading too many issues of CIO magazine between rounds of golf!

    Either that or this was a good troll. Either way... cloud is never cheaper for non trivial deployments and work loads. Period. If your cloud service is cheaper for real work then you're seriously incompetent and going out of business soon anyway.

    Who is taking the 3am call when shit breaks? Your developers? They checked out at 3:45pm, went to the bar, did the deploy to production from there, got super drunk and passed out.

    NoOps: more silly shit from people who never worked in the real world.

    Too funny!

  14. Is this true? by phantomfive · · Score: 1
    Here is a quote from the author:

    Now your customers say, 'where's your mobile app? I want to be able to have voice assistance to talk to your particular product.'

    Is that really true? I thought apps were mostly a dead end now, and everyone is just writing for the web. Do people really want voice recognition in apps?

    --
    "First they came for the slanderers and i said nothing."
    1. Re:Is this true? by 110010001000 · · Score: 1

      No. Companies like Google WISH people wanted voice recognition. It is another data point they can collect and track. Voice recognition is like VR: a technology that no one wants.

    2. Re:Is this true? by Anonymous Coward · · Score: 0

      It's mostly just a human centipede. The overlords tell people that apps are the source of everything great, people demand apps, the overlords say, "hey look, they're still demanding apps", it never goes away. They just keep adding new "revenue streams" to milk the chattle, the old ones never die. This works because debt can be passed from one generation to the next. This is how you end up with people born into negative wealth.

    3. Re:Is this true? by JustNiz · · Score: 1

      Meh, writing for the web is soooo 10 minutes ago..

    4. Re:Is this true? by Junta · · Score: 1

      I want and use my VR. I'm very sad if it fails to take off more than it has.

      Voice recognition on the other hand.. Meh

      --
      XML is like violence. If it doesn't solve the problem, use more.
  15. By Neruos by Anonymous Coward · · Score: 1

    "With Amazon Lambda and other microservices, you just need HR to hand out IAM accounts, and a company really doesn't need an IT staff whatsoever. Just some CI/CD mechanism to get pushes in production, and that is basically it.

    Ops is dead. Who needs to rack and stack physical servers when the cloud takes care of that, and far cheaper. Who needs OS guys, app guys, net admins, and DBAs when serverless services replace all this?

    Lets be real... the future is NoOps. Pay your Amazon bill, and they take care of your IT infrastructure."

    You sir, are a moron. This is the crap that PMO/MBA wannabe's push to C suits to get into Director/VP promotions and once it's done, they bolt. Anyone who says DevOps or Cloud as a term to solve a companies problems is going to blackhole the company. Unless you're a mobile app provider with 1-2 products, you don't do that. All Fortune 100 companies user Ops and even some Cloud and some DevOps, but in percentages where they work.

    Next person that says web-scale......

  16. Re:The problem is that there is no 'Test' in DevOp by Anonymous Coward · · Score: 0

    Unless you go the Netflix route and do tests in production.

  17. Re:Yawn by Anonymous Coward · · Score: 1

    I wouldn't hire someone stupid enough to publicly say what you're saying here to work on my team. Good sense is important in any developer.

  18. AKA Permanent Instability! by Anonymous Coward · · Score: 0

    A certain generation and culture of developers always praise continuous delivery to the heavens. Implying that *every* commit should be stable, bug-free, extensively tested and released.

    But on every single project I've seen, it results in the opposite: Permanent instability.
    No version is allowed to mature anymore. None!
    Yet everyone involved can't admit to having been wrong, and so the whole thing turns into another Cult Of (Riding The) Dead Horse.

    And for everyone with a bit of brains, it is blatantly obvious why that is: The snapshots *aren't* tested to an acceptable level, in the real world. Because the next one would be finished earlier, or if blocked, development would slow to a crawl.
    In the time you have, you can only test *some* snapshots. And if you care about stability, you can only release those.
    Continuous delivery and stable software are mutually exclusive.

    1. Re:AKA Permanent Instability! by Junta · · Score: 1

      I think it's the efffects of the bubble. The late 90s did have whiffs of this too, though that bubble burst earlier in the cycle and communication wasn't as rapid as it is today, so it's a bit more severe. Developers in general had a wake up call after the bubble burst last time and things settled back down for a while there into more manageable offerings.

      The real root cause is not the fads and rhetoric, those are a combination of band aids and denial about the reality that the market is flooded with students that became programmers because of big dollarsigns in their eyes and not so much passion for the industry, and companies have a very hard time recognizing the value of the more intrinsically dedicated and focus more on X developers should be able to do this, and if not there must be a magic process to fix that'

      --
      XML is like violence. If it doesn't solve the problem, use more.
  19. Re:Yawn by Anonymous Coward · · Score: 0

    I wouldn't hire someone stupid enough to publicly say what you're saying here to work on my team. Good sense is important in any developer.

    For all you know, lucm is working for you right now.

  20. I was actually waiting for this. . . by Salgak1 · · Score: 1

    . . . because, sooner or later, DevOps HAD to no longer be the New Hotness. . . Of course, the question is. . . what comes next ?

    1. Re:I was actually waiting for this. . . by ChatHuant · · Score: 1

      sooner or later, DevOps HAD to no longer be the New Hotness. . . Of course, the question is. . . what comes next ?

      I think the time has come for MangeOps - A software engineering practice that aims at unifying management (Mange) and software operation (Ops)

      The goals of MangeOps span the entire business pipeline. They include:

      • Improved customer complaints frequency
      • Faster time to failure
      • Higher failure rate of new releases
      • Shortened lead time between critical breakages
      • Faster mean time to market speak recovery (in the event of a new release crashing or otherwise disabling the current system)

      MangeOps aims to minimize the predictability, efficiency, security, and maintainability of operational processes. MangeOps principles demand strong interdepartmental time wasting - team-building and other employee engagement activities are often used, to create an environment that reduces communication and forces frequent and mandatory cultural change within an organization. Teamâ"building activities can include board games, trust activities, employee engagement seminars and other activities that displace engineering time.

    2. Re:I was actually waiting for this. . . by nasch · · Score: 1

      The only thing I saw in there that I liked is playing board games at work.

  21. Re:The problem is that there is no 'Test' in DevOp by Junta · · Score: 1

    I think the complaint is developer written tests of their own code is frequently considered a replacement for an actual QA team, whether unit or integration test.

    One of the *massive* gaps is if the developer is batshit crazy and designs horribly because they are not the target audience and do not understand the target audience, the tests are going to sail on through and be horrible because no 'mere mortal' could check that developers bizarre vision. Also a developer can be plain dumb and think the wrong thing and make tests that only will pass when a broken implementation is thrown at them.

    The biggest problem is the buzz of CI/CD brands developers in general as infallible. CI at least is a good idea in and of itself to augment a process, but the collateral damage of how those words are interpreted by management is severe.

    --
    XML is like violence. If it doesn't solve the problem, use more.
  22. Re:The problem is that there is no 'Test' in DevOp by john.sheley · · Score: 1

    While I agree with you that good test automation is challenging, I would say that it is no worse than on-premise product development. Effective testing requires engineering commitment to make test automation development co-equal with product development. It is an immense challenge. BUT, having worked CI/CD for both an on-premise product and a cloud-based (AWS) product, cloud seemed easier. If for no other reason than it is much easier to manage the equipment.

  23. Bad Headline by Anonymous Coward · · Score: 0

    He says the battle is over and DevOps won and people need to move on to the next thing, not that people need to 'move on' from it in the sense of rejecting it or forgetting about it.

  24. Continuous masturbation by Anonymous Coward · · Score: 0

    This was my internal code name for that thingy, after it married the rampant bureaucracy at $WORKPLACE and begat a monster.

    Before, I'd been doing it all the time, but the code name was different:

        configure && make && sudo make install

    This one Just Worked (TM).

    The moral of the story: it's not the tools, but how you use them. Don't buy a $24 hammer if you can't use a $2.49 hammer.

  25. Re:The problem is that there is no 'Test' in DevOp by zapadnik · · Score: 0

    So use System Tests as well in your Continuous Integration setup. Have you never heard of them ?

  26. Re: The future is NoOps -insightful? Lololol by Cederic · · Score: 1

    Not strictly true. If you have a globally applicable product that must nonetheless comply with local data residency and access rules then cloud services let you deploy globally without needing a complex fragmented physical estate.

    For regions with sizeable footprints there will be an inflexion point at which self hosting makes commercial sense. Even there I'd constrain the solutions to the same deployments used in the cloud, for dev and ops simplicity.

  27. Some Definitions by Anonymous Coward · · Score: 0

    CI: Continuous Integration (in this context)
    CD: Continuous Delivery
    CD: Continuous Deployment
    (Yes, I, too would like to punch whoever picked a second word that starts with D; it's probably the same guy who gave us Verification/Validation and Authentication/Authorization.)

    Continuous Integration means not having code hanging out on an isolated branch for more than a day or so: instead, it is continually integrated into the main branch (master, trunk, etc.). People started calling automated build servers "CI Servers" because they're very helpful in supporting CI, but that means many people now think "CI is what CI servers do," and so someone had to go and invent "Trunk-Based Development" as a term for actual Continuous Integration.

    Continuous Delivery means very frequently giving your organization a system that's of high-enough quality that they could deploy to production if they wanted to. Dave Farley and Jez Humble wrote a book on this; go read it.

    Continuous Deployment goes one step further and actually deploys that system to production. This turns out not to be crazy if you pay a lot of attention to quality in your pipeline.

    Human QA is still a valuable part of these pipelines, as is actual customer involvement. Automated testing is a way to take the load of repetitive testing off the shoulders of Devs and Testers, so they can do things that require actual human thought.

    Yes, I know your dysfunctional organization screwed these things up horribly. Consider getting another job instead of bitching about good ideas.

  28. Is gmail out of beta yet? by Anonymous Coward · · Score: 0

    Is gmail out of beta yet?
    Or do they still have 3 outages a year?

  29. Agree by Anonymous Coward · · Score: 0

    Agree.. Its time we stop coddling older devs who still live in the days of siloed responsibilities and push them out of the best. Devs who can't be cross functional arent really devs.