Sam Stone believes Trump's tweets

The BBQ Pit
1424

Yes, we get it. You are an awesome programmer, and you need to show everyone that you know what a code review is. Good for you! If I had a lollypop, I’d give you one.

In the meantime, I was answering the people who were saying that you can’t learn about a progrmmer by looking at a few screens of code and a checkin list. I hope you weren’t one of them. I thought some people might actually like some examples from a code review. Good thing you were around to put me in my place for committing an act of explanation.

I’ve never been pitted before for a straitforward, non-political factual post explaining a technical issue some people might be interested in. So congratulations on that. We don’t want people around here doing something crazy like fighting ignorance.

Yeah, we wouldn’t want anyone on the SDMB ‘crafting a narrative’ on tiny scraps of fact.

Hey, has Trump been charged yet over the documents the FBI confiscated? Did he sell them to the Russians? How many posts were in the Trump FBI document thead describing what Trump probably did before any information was released?

How about the Mueller investigation? How much time did Trump do for the obvious crimes listed in the gigantic Mueller investigation thread?

Oh, did you hear that the Hunter Biden laptop is obvious disinformation and that the computer store owner is an obvious Republican plant? I read it right here on the SDMB. Some guy even got pitted for suggesting that the laptop might actually be a real issue. Good thing you fought ignorance and set him straight with insults without any ‘narrtive building’ at all, while you were awash in all those facts.

You people are a joke. The one thing that makes this endless, omnibus ‘get Sam Stone’ thread tolerable is that I spend about as much time laughing at people as I do being annoyed.

When we were debating whether the Pit should be closed, the claim was made that the pit was necessary to,combat racism, sexism, misogyny and homophobia. I guess we should add ‘describing a code review’ and ‘building a narrative I don’t like’ to the list.

1425

No you’re being pitted for a lot of extremely dumb opinions on why there’s a perfectly logical explanation for Musk’s last several weeks of publicly lighting money on fire.

1426

Or more properly, building a fantasy narrative built upon stuff you extracted from your rectum.

1427

I provided potential alternative explanations for some of this stuff. I backed them up with facts and explanations, while saying that it’s also possible that what he’s doing is a shit-show, because none of us were actually there and know what went on. As far as I can tell, I’m just about the only one going, “Wait a minute - it could be something else. It might not be, but it might”.

What I’m really getting pitted for here is not ‘building a narrative with minimal facts’, but for adding a note of caution to YOUR narrative being built without any facts.

When the pitfork-wielding mob forms, no one is hated more than the person who says, “Wait… is there a chance she’s NOT a witch? You can interpret her actions in another way…”

And the response from the crowd is, “Only a WITCH would defend a witch!”

1428

Your ability to find small pieces of peppermint candy in large piles of horseshit truly astounds me.

1429

He has long since lost the ability to smell horseshit. Thus the peppermint comes through clearly.

1430

Well, thanks for the backhanded compliment but in fact I’m not all that much of a programmer, especially now that most of my coding consists of writing Python/NumPy/SciPy applications for almost exclusively my own use with a little bit of Matlab and the occasional Fortran 77 mod. However, I don’t really have to be a rock star coder to see that you were clearly trying to ‘school’ @wolfpup in your superior knowledge of what a code review consists of, and while the examples you gave of the “best practices” variety are certainly part of a code review, they aren’t really the main focus of the effort, which is more about ensuring that the code is written to internal standards and practices, that there are adequate unit/baseline tests, that commenting is done in a consistent style, that static analysis and safety/vulnerability testing (where applicable) has been performed, dead code is excised, et cetera. A big portion of code review isn’t even about the code per se but about verifying requirement flowdown and reviewing documentation, and is intended to convey knowledge across the team so it isn’t just the guys writing the code that understand how it works.

Insofar as you can even evaluate the expertise of an individual programmer by looking at code snippets (and in a large organization like Twitter any particular unit of code is likely to have gone through several hands, especially if they are doing ‘Agile’-type development), a reviewer with no prior knowledge of the internal standards, framework, or application isn’t really going to have any real insight into how good of quality the code is aside from the superficial issues that you list. And doing a code review is about the worst way to evaluate the performance of individual contributors. Giving employees coding tests, asking them about applicable standards and practices, and providing hypothetical logic problems pertaining to typical coding issues would be far more straightforward and insightful regarding individual performance and ability. But that also takes effort and forethought which are qualities solely lacking in this ‘management transition’.

Stranger

1431

You are the Joke.

I never doubted that the laptop belonged to Hunter Biden. He as much as admitted it was probably his, before the election. The issue, and the reason it was considered disinformation, was it had been in the hands of agents of a country that we have an adversarial relationship with. It’s a chain of custody issue, and the fact that Guiliani and friends lied about the origin also makes it suspect.

I also pointed early on that the unhinged republicans could have Hunter’s laptop, but as they showed with the “Climate gate” emails and Hillary Clinton ones, the demonstrated bullshit that conservative sources used to mislead many should have been enough to discourage anyone with common sense from continuing to use them as sources. (Me and other Dopers and right wing sources being right about the laptop being real was not the reason why one should trust the right wing about how they spin the contents of the laptop)

Again, it is clear by now that you willfully ignore that the point about the laptop being real was granted by many, the reality is that you don’t want to deal with what is going on, the evil efforts from many conservative sources to mislead their viewers and readers about what the contents of the laptop really are.

1432

I know nothing about this, so my question is, are code reviews generally conducted using screenshots and printouts of code? Do code reviews generally require the coders to bring those screenshots or printouts?

Because, and I reiterate I know nothing about this, but it seems to me that Twitter already has their code. They shouldn’t need the programmers to show up with it; they should be able to pull it up on their screens at will.

I’m assuming that Twitter tracks which engineer modifies or creates which code; is this a correct assumption? It’d astonish me if they don’t know who wrote which code unless folks show up with screenshots, but for the third time I know nothing about how this works.

1433

No and no.

Yep.

That is correct.

1434

In a Dilbert cartoon or a Mike Judge film, perhaps. In a modern software application or web platform development organization, it is a rational assumption that production code is tracked by checksum, stored in a controlled repository system, and should be available to anyone in the organization who has authorization to access the repository. Now, “the code” in development is far more complex than a single application stack; there may be many different branches of code developing different capabilities and in various states of maturity and verification & validation cycles, so figuring out what is in ‘production’ versus some state of development is itself a non-trivial task, and just because some bit of code has been “submitted” doesn’t mean that it is ready to ‘go live’; a development team submitting code may be the first step in many stages of V&V, quality assurance, independent code review, strategic review, et cetera.

Tracking which programmer or software engineer is responsible for any particular piece of code is, while not impossible, certainly complicated by the fact that most modern enterprise software development paradigms have many people touching every piece of code. You can see who did a submit on some bit of code, and in a (presumably) mature company like Twitter you’ll have an entire documented chain of review and acceptance from the Software Requirements Specification (SRS) flows all the way to deployment of living code, but there are many cooks in the kitchen and figuring out which one under/oversalted the stew is not an easy task of just looking through the submit history, notwithstanding that a lot of ‘new’ code is actually highly reused, either by deliberate object/framework reuse for development efficiency to ‘copypasta’ with zero effective traceability.

Anyone asking for screenshots or printouts might as well be asking for the programmer to bring in his input deck of punchcards. Twenty-odd years ago when I was doing Java development on tutorial software for mechanics instruction, I maintained a record of paper printouts because we were often running out of disk space on individual workstations and/or Unix accounts trying to manually do revision control (this was in the early days of concurrent revision control and none of us were clued in enough to set up a CVS repository) and I quickly filled up multiple bankers boxes of printouts that, while useful for our small team and the system we had for tracing changes, would have been totally worthless to anyone coming in trying to review the program development. And, of course, such paper records have absolutely no traceability; I could print out anything and claim that I’d done all of the work, and who is to say otherwise?

Stranger

1435

As I pointed to Sam already in the Musk buying Twitter thread, a lot of what Sam has posted could be valid, but mostly to deal with companies where it is clear that there are issues with the programmers or engineers, as when one encounters a product that is not performing as promised. Musk took over a company where, clearly, programmers and engineers were already delivering actual solutions to problems that were unacceptable ten years ago. There are still issues, but keeping it mostly reliable, with constantly changing algorithms, has been a noticeable feat.

1436

What you’re being pitted for in this thread started with believing that Trump’s tweets were something more than meaningless gibberish. You still haven’t accepted and admitted that you were foolish to do so.

1437

Or a high school level programming class.

Pretty much everything that @Sam_Stone laid out as examples of bad code are things I learned to avoid in junior year in Pascal. If people are making those mistakes, they probably should be let go, but I also don’t think that there are going to be many who make those mistakes who have actually gotten the job. I’m certainly no professional programmer, but I would avoid those.

Musk’s code review may have been appropriate if he were a high school teacher grading a couple hundred line code project.

It’s like evaluating a book based on its spelling. Sure, spelling is important, and lots of spelling errors could indicate something about the competence of the author, but it tells you nothing about their ability to create a workable plot.

1438

AKA, a fantasy narrative.

1439

If one were to give Sam the benefit of doubt and take him at his word he’s just playing devil’s advocate. The problem is a) that accepting this requires to ignore all the evidence that he’s often emotionally invested in being correct to a level unlikely for someone not considering the side being argued correct, b) a lot of the arguments require one to ignore all previous evidence.

Like this code review debacle. He’s actually spent the time writing a lengthy post on the most basic bad code imaginable, as if people don’t grasp the concept of bad code and are actually judging Elon’s move based on the full context of happening in a company that’s been able to hire selectively for several years and being in the form of “send me screenshots of the best code you’ve written lately”. Anyone not going to sleep staring at the Musk poster on their ceiling would say that looking for those kinds of errors in that fashion would be a poor use of time and resources.

1440

I know this might sound crazy, but there are people on the SDMB who aren’t even programmers. I know, hard to believe. Perhaps they might like an example of what bad code looks like.

I love it that a straightforward example of what we are talking about has become a ‘debacle’ because some of the mean girls in the pit took exception.

It’s rather pathetic.

1441

What is pathetic is how you misrepresent what others have said in order to make yourself the victim.

It wasn’t that you gave an example of what bad code looks like, it’s that you made the assertion that Musk was weeding out people who apparently failed high school programming.

Sure, someone has a bunch of crap in their code like you put, and you ask how they managed to even get hired in the first place. But you don’t go looking for it assuming that the engineers that chose to stay are shitty programmers.

That’s the part that you seem to refuse to understand here.

1442

I started writing a lengthy post, from the perspective of someone in the software world, as to why Sam’s points are misleading. But why bother? He’s not interested, not really, or he’d actually engage with the points other people are making rather than parroting the same nonsense.

1443

For people who aren’t programmers, how is showing them examples of bad coding going to mean anything to them, any more than correcting bad French grammar going to be useful for someone not fluent in French? You clearly concocted this example to show off at how much more knowledgable you are than everyone else about coding practices and thus what Musk’s intent was, and incidentally demonstrated that you don’t actually understand the purpose and utility of code or peer review.

If you imagine all of @Sam_Stone’s posts as being read by Jordan Peterson is becomes very clear. He’s just fine with attacking others or trying to assert his superior knowledge of some principle but when caught in an error or misapprehension he goes into full-on victim mode, whinging about the confederacy of dunces is conspiring against him.

Stranger