Monday, 4 November 2013

09:44 – With all of the news about the catastrophic rollout of Obamacare, I’m surprised that no one has pointed out that these problems lose Obamacare the fig-leaf of Constitutionality that the contingent SCOTUS ruling provided. It ruled that Obamacare was Constitutional based on representations that are clearly false and were known to be false when they were made, and it made clear that its ruling was based on those representations being true.


16 thoughts on “Monday, 4 November 2013”

  1. Developers just cannot leave a good thing alone. Sometimes I have a dozen or more tabs open in Firefox. While researching something, I end up passing over the best thing in the bunch of tabs, and need to go back to find which tab out of the dozen it was in.

    Formerly, I could enter a word or so into the search function, then click on the other tabs and hit Ctrl-G until it showed me what I was looking for. With the latest FF upgrade, when I hit Ctrl-G in a different tab, it now shows me a blank search dialog, instead of just searching for the last string I entered.

    So much was promised for computers and what little is delivered is always changing.

  2. Developers just cannot leave a good thing alone.

    Hey, as a software developer, I resemble that!

    The estimate to write debugged code is 10 lines per person per day.

    I have written well over 1,000 lines of working (next step beyond debugged as it actually meets the requirements) C++ code in a day. Of course, the code was task oriented and fairly boiler plate.

  3. Code productivity varies hugely. First, by programmer: A good programmer is at least 20 times as productive as a bad one. More: the good programmer can implement concepts that the bad programmer will never understand. Then by project. Working alone is vastly better than working in a large team. User interface code eats time; business logic is much faster (and more fun) to implement.

    My best was 10,000 lines of highly complex Prolog code in about 3 months, including testing and documentation. But I was working alone, there was no GUI, and it was in a domain (logic systems) that I had just completed my doctorate in.

    Finally, lines of code are not lines of code. I got a student project last Spring that spent nearly 10,000 lines of code just creating a single window. A good programmer could have done the same task in less than 500 lines. They had just copied-and-pasted code dozens of times, for every single little control, rather than using classes and methods to do each job once. Horrible stuff…

  4. Finally, lines of code are not lines of code. I got a student project last Spring that spent nearly 10,000 lines of code just creating a single window. A good programmer could have done the same task in less than 500 lines. They had just copied-and-pasted code dozens of times, for every single little control, rather than using classes and methods to do each job once. Horrible stuff…

    I think the problem with the affordable care act goes deeper than a coding problem. I think there are design issues. We know at this point that there are testing issues. If there are 55 subcontractors working on this without a prime contractor, it’s hopeless.

    In my opinion, the 55 contractors who worked on this project all made the same huge mistake. That mistake is they submitted a bid on the project.

  5. During my active years working fulltime in television, I never committed to my superiors how long it would take to do anything. They did not know, and I was not about to tell them. They loved live or ‘live on tape’ programming, because to them, it used only about an hour’s resources for a half-hour program. We on the line, called them ‘quick and dirty’ because none of those shows ever won awards. It was the programs with lots of prep, shooting and editing of roll-ins, and/or documentaries that won the awards. We all hated the quick and dirty stuff, which almost always relied on a host to make or break the show. There were a lot more breaks than makes.

    The only thing my bosses could quantify about my work was my shooting ratio. They kept track of that for every program maker. Mine was never over 12:1 and often around 8:1. Other producers started at 15:1 and went up to as much as 50:1. Management got concerned when you hit over 30:1. They never could figure out why mine were so low, but yet I still won lots of awards for them. The key was knowing when you ‘had it’. A lot of producers and directors never know if they really have a good take until they see it in the edit suite. Early on, I learned how to tell in the field. If I knew the take was good, I did one more for safety sake (only when working at the top levels can you rely on some hitch not to get you when you look at it back at the studio), and moved on.

    When Siskel and Ebert first started, the studio taping stretched over 2 days for a half-hour show. When I took over, it was never more than 4 hours, and most often just 2 to 3. My goal was never to have to do a segment twice because of crew faults; if we did it again, it was because one of the two guys wanted to do it over. The crew pretty much never screwed up, and if we did, I stopped the take so we did not continue doing something that was known to be bad. A lot of people continue and call it a “rehearsal”.

    I have been told by people who have managed both, that coding software is a very similar environment to making TV shows and movies. Both are bottom-up creative processes, and almost always fail when managed as top-down. My daughter in-law the software programmer, says she is—these days—constantly faced with bosses who ask her how long it will take to do some coding, then clutching her answer in hand, confront her and ask, “How can we cut this time in half?” She has stopped telling the truth about how long things will take, but she still tells them nothing can cut the time in half.

  6. I was eating lunch and heard that first hour of Rush’s show. There is no way this regime could not have known that there would be a major shift in the whole healthcare field by this legislation. Rush told of a woman being treated for stage 4 cancer who can’t keep her doctor, nor can she keep her health insurance (they cancelled her). No way can she keep going with the same system of treatment that has so far been effective in keeping her alive and has already paid out over a million dollars to care for her.

    What it takes for insurrection to start in this country is beyond me, but if this does not do it, nothing will and Barry will have his way with us.

  7. My daughter in-law the software programmer, says she is—these days—constantly faced with bosses who ask her how long it will take to do some coding, then clutching her answer in hand, confront her and ask, “How can we cut this time in half?” She has stopped telling the truth about how long things will take, but she still tells them nothing can cut the time in half.

    She should give them a copy of “The mythical man month”:
    http://www.amazon.com/The-Mythical-Man-Month-Engineering-Anniversary/dp/0201835959/

    After all, if one woman can have a baby in nine months, can’t nine women have a baby in one month?

    I always estimate time in weeks or months, double it and give that answer. That way, I am usually not under by more than 50%. I should really quadruple my estimates instead of doubling them. I did a project last year where I estimated one month to implement and gave two months to the sales guy. The project took me four months calender time with my leadership duties and emergency bug fixes interspersed throughout that time.

    My BOD wants a time estimate to recode our Windows user interface (850,000 lines of C++ code married to the Win32 API with over 150 dialogs and a custom graphical user interface) into a internet browser. I’ve been telling them 20 man years and getting a very emphatic “no way it is that hard” in return. Sigh. And they know better.

  8. Finally, lines of code are not lines of code. I got a student project last Spring that spent nearly 10,000 lines of code just creating a single window. A good programmer could have done the same task in less than 500 lines. They had just copied-and-pasted code dozens of times, for every single little control, rather than using classes and methods to do each job once. Horrible stuff…

    This is how I believe that the Obamacare website was coded. Scary stuff.

  9. “What it takes for insurrection to start in this country is beyond me”

    What will be the battle lines for the insurection? I don’t think it will be by state or region. Nor the rich vs the poor. I keep thinking the coming battles (figuratively and literally, and hopefully years away) will be the entitled and non-entitled. Those who receive a monthly kiss from the government for doing little or nothing vs those who work so hard to support themselves and have most of it taken away. That, of course, means the front lines will be in every major city in the US.

  10. Jim—good to see you back. Hope the trip went well. That t-shirt is a knock-out. Only problem is that my conservative friends do not see anything wrong with the NSA. And on the other hand, neither do the liberals. So the NSA is good to go in the USA.

  11. “the 55 contractors who worked on this project all made the same huge mistake. That mistake is they submitted a bid on the project”

    Are you kidding? What a gold mine!

    Seriously, if you are in it for the money, and don’t care about the actual project results, a high-pressure contract like this has profit written all over it.

    Specifically I would bet that a lot of the contracts were “cost plus”. This happens a lot for projects under time pressure, where the contractors are not able to give a believable estimate of expenses. “Cost plus” lets them bill whatever their costs happen to be, plus an agreed profit margin. Otherwise known as the gravy train…

    That said, Chucks DiL is right: there is as yet no way to reliably estimate a software project, unless it is something really repetitive like “configure another webshop just like the last one”. I’ve never met a software type whose initial estimates were too high.

    My best method is to spend an hour or two analyzing the project, come up with an estimate, and multiply by three. @Lynn: You say you double and are under by 50% – see, my highly disciplined method is better! 🙂

  12. I need to quadruple my time estimates. Gotta leave time for staying up until 3am last night reading “Captain Vorpatril’s Alliance”.

  13. “What it takes for insurrection to start in this country is beyond me, but if this does not do it, nothing will and Barry will have his way with us.”

    Not this, that’s for sure. A few thousand anecdotal medical treatment/insurance horror stories won’t do it. As I grow tired of saying, it won’t happen until a critical mass of the population is actually hurting here; when the lights go out and the store shelves are empty. After several weeks, shit will start to hit the fan.

    A preview of sorts may be coming up before Christmas; in another couple of weeks tens of millions of food stamp recipients are going to discover that their monthly allowance has been significantly cut, as of November 1. And our overlords and masters are planning a simulated bank crisis of some sort and a Grid-down “drill,” probably to keep stoking the fear and anger deliberately.

    We’re looking at an increasing energy crisis, no matter what miraculous enterprises are beavering away in our western states and the Canadian provinces; increasing ethno-racial strife, which the globalists deliberately foment and incite; and eventually, the secession-by-default of the various blocks of states and provinces. Not tomorrow or the next day or next year but over the next fifteen to twenty years. This gigantic house of cards is simply not sustainable, by any reasonable measure of basic arithmetic, which even I can understand.

Comments are closed.