Hub of Reason

nomen agentis

00:00

@Oddthinking Are people actually recommending vinegar over creme cleanser? or are they just saying that vinegar is effective

a good denialist, after all, puts words into the mouths of opponents :)

Oddthinking

@nomenagentis How much is Big Vinegar paying you to be a shill?

2

nomen agentis

05:14

A lot of vinegar

DampeS8N

13:29

Did discussion about skeptics.stackexchange.com/questions/27590/… leak over here at all?

nomen agentis

No but it should. The comments are an awful place to discuss

DampeS8N

@nomenagentis Agreed

I was hoping it had for that reason. I don't like seeing flusterclucks like that on this SE.

well, any SE

but especially this one

nomen agentis

I've flagged comments that stray into discussion.. When mods wake up I expect a notice to move discussion to chat

DampeS8N

Even more disheartening when it is fueled by mods and high rep users. Is there anything we can do to clean it up?

Probably correct

But also probably too late.

nomen agentis

They often fully wipe comments

DampeS8N

13:35

Wiping the comments will not solve the underlying issue. We can start a meta about what to do when a question sparks two divergent theories of correct answers.

In this case the nuances of what a shortage means

nomen agentis

I don't find the question interesting especially after StackExchange clarified they were really saying there is a shortage of people that meet their standards for agile developers

DampeS8N

hm

Do you think it is unlikely for other questions to crop up with the same problem of definitions and that are also interesting to answer?

nomen agentis

Yeah I've seen a bunch

My position is that if an element of the question is vague, but if it is vague because the original claim was vague rather than the asker introducing vagueness, then so be it

DampeS8N

I'm sure a solution can be found that will prevent this sort of thing. Or perhaps given time it resolves itself.

hm

nomen agentis

A good answer will look to how experts or preferably the original source of the claim meant to use the vague term

And base an answer around that, but clearly stating their assumptions about what the vague term meant

If the term is so vague that it will attract a flurry of different but all correct answers, then it is maybe too broad

I don't know how StackExchange deals with it though.. I'm sort of new

DampeS8N

13:42

In this case I am more concerned with how we can get our mods to act more appropriately. I think probably this question should have been closed with directions for the asker to clarify the salient point.

I used to mod Science Fiction and Fantasy SE, so I know something about it.

nomen agentis

Well, in this case, the claim itself has was vague.. I don't know what StackExchange meant by shortage, other than what they say in their clarification and the link that they pointed to to support that claim

I don't think its fair to ask an asker to impute their own definition for that word.

I haven't read every comment, but what are mods doing poorly?

DampeS8N

@Sklivvz and @Oddthinking Do you guys think it is the right action if this comes up again to close the question with direction for the OP to clarify or select between two interpretations? Or would offloading discussion to chat be sufficient. This is your rodeo.

nomen agentis

I guess sklivvz did start talking about what in his opinion was the "real" claim

Which is perhaps hijacking the question. The claim is the claim... Just because another rewording would be more interesting to a mod doesn't make that version more "real"

Oddthinking

@DampeS8N Rodeo? I have a rodeo!

DampeS8N

@nomenagentis Mods can accidentally hijack questions just by virtue of having the diamond. It gives unintended authority to... YEEE HAAW @Oddthinking

nomen agentis

13:48

Odd thinking did this skeptics.stackexchange.com/questions/27590/…

So some clean up is happenin

Oddthinking

@DampeS8N Yeah, I would be nice to be able to put down the star and the gun sometimes, when we just intend to take a personal view, not the official word.

nomen agentis

Ya I like how Wikipedia doesn't identify administratora

Oddthinking

I am pondering the discussion here that we should close the question as Unclear.

nomen agentis

*s

DampeS8N

@Oddthinking It seems appropriate to me to have done that from the start. I haven't seen one get this out of hand in a while. And some simple clarification prodding, which is what closing can do for a question, could have gone a long way.

nomen agentis

13:51

But would you ask the asker to clarify? Even though that's not really possible to do since the StackExchange claim was itself vague? Or would you ask the asker to find a more specific version of the claim? And actually, if notability is all that needs to be met, then almost any more specific understanding of the claim would be notable... Enough people believe all different forms of the shortage claim

DampeS8N

I know there is a risk of driving the OP off with closing, but when combined with a comment stating why it was closed and asking for clarification that rarely happens.

@nomenagentis Claims don't define the criteria for acceptance, the asker does.

Oddthinking

@DampeS8N Always the dilemma. Trigger finger on the preventative closure (which leads to annoyed OPs) or just a few comments. I think the mods here have a reputation for trigger fingers here, and this sort of question is the reason.

nomen agentis

At this site, as far as I can tell, the question has to be challenging a claim

And askers can't delimit the evidence that answers can bring

Oddthinking

@nomenagentis This might be the root cause - there are multiple believed claims here.

DampeS8N

@Oddthinking It is only a problem if the question stays closed.

If after 24 hours the OP is a no show or seems to have moved on, mods can and should reword the question themselves for the clarity they think is needed.

Oddthinking

13:56

FYI: Closed now, but don't let that stifle the debate. I want to see it clarified and re-opened because it has struck a nerve.

DampeS8N

@Oddthinking That's true. Even stating both and asking for answers to address both possible meanings would have been helpful. That might even make the question more helpful overall

nomen agentis

I don't get how it struck a nerve haha

Its just a question about (in my opinion) demographics

Oddthinking

@nomenagentis Every now and again we get a question that affects software developers and we are reminded where our roots are from!

190

Q: Do bigger or more monitors increase productivity?

There is a ton of anecdotal evidence that adding more screen estate (more or larger monitors) leads to an increase in productivity, especially for programmers. For example, where I work, all programmers have either two 20" screens or a single 24" screen. There is no agreement on which is the opt...

nomen agentis

I guess there are lots of developers around!

Oddthinking

178

Q: Has Stack Overflow saved billions of dollars in programmer productivity?

John Carmack is a renowned developer and CTO. His Twitter account has over 100,000 followers. On the occasion of Stack Exchange's 5th anniversary, he quipped on Twitter: [... Stack Overflow] has probably added billions of dollars of value to the world in increased programmer productivity. ...

economics programming-productivity

nomen agentis

13:57

Oh this is the "painting a shed" problem

Oddthinking

122

Q: Is closed-source code more secure than open-source?

My computing teacher told us that closed source software is more secure than open source software, because with open source "anyone can modify it and put stuff in." This is why they do not want to use open source alternatives for learning to program, such as FreePascal (currently using Embarcader...

security software

nomen agentis

If you've heard of that

DampeS8N

@nomenagentis There is a problem in the field of programming where there are a huge portion of the work force that is barely competent or incompetent and so when you are looking for candidates for a position you end up sifting through 100 applicants to find 1 candidate that can solve trivial programming problems.

The analogy of asking a carpenter which end of a saw to use is not even hyperbolic.

nomen agentis

I read an article recently that said the bimodal distribution of programmer skill is a myth

And that it follows a bell curve like other skills

But perhaps amongst the people still looking for work, that is a different distrubution

DampeS8N

@nomenagentis It certainly does follow a bell curve, but the problem is that for anything beyond basic needs, which is what most people are highering for, you need people off the lip of the bell and not the bulge

nomen agentis

14:00

Oh, so the bell is there but the threshold for acceptability is at like the 90% percentile?

Oddthinking

So, the question quickly becomes a "True Scotsman Fallacy"

"There is a shortage of programmers."
"What about these unemployed ones?"
"Oh, there is a shortage of REAL programmers."

3

DampeS8N

@Oddthinking You've got it baby

nomen agentis

Ha exactly

Oddthinking

On the other hand, the "N positions of every developer" claim is really "There are N people who aren't serious enough about development to pay the right price, and offer an attractive environment.

nomen agentis

That was hard to notice in this one

DampeS8N

14:02

In the absence of a specific measure of quality, that's exactly the problem.

nomen agentis

Yeah no "operational definition"

en.m.wikipedia.org/wiki/Parkinson%27s_law_of_triviality

Oddthinking

I think there is a severe shortage of top-end DSLR cameras, because I can't buy one for under $100.

nomen agentis

That says a bit about why every programmer has an opinion on this one

DampeS8N

And the deciding factors for what defines this particular Scotsman is a host of skills like problem solving, communication skills, being a self starter. All the BS marketing terms people like to put on their resume to pad it out, only it is people who actually can do those things.

the big one is problem solving

Oddthinking

So, this (the loose definition, and the bikeshedding) is why I call to the economists and say "Hey, you guys have probably thought this through. When is it a "shortage"? What other term should we use?"

nomen agentis

14:05

Yeah that's what I would prefer we do... Look to how experts analyze this problem

DampeS8N

To be the lone guy keeping a website up and running in a corporate environment requires a lot of skills, chief among them being able to think through a problem and find a solution and actually perform the tasks needed to get it solved. Often with only duct tape, an old PC and no budget for tools, hardware or hosting.

nomen agentis

And being an expert programmer doesn't make people an expert in analyzing programmer shortages

Oddthinking

Shortage

A shortage is a disparity between the demand for a product or service and its supply in a market. Specifically, a shortage occurs when there is excess demand; therefore, it is the opposite of a surplus Shortages are related to price--when it is "too low," there will be a shortage. In most cases, a shortage will compel a firm to increase the price of a product until it reaches market equilibrium. Sometimes, however, external forces sometimes cause more permanent shortages, in other words, it is a situation in which the quantity supplied of some good or service is less than what it would be if the...

I tried to link to #LabourShortage, further down the page.

DampeS8N

@nomenagentis The problem is that each environment is different and each set of problems novel. So a generic ability to think your way through any problem is the big ability that these candidates lack. That's what fizzbuzz does. It shows the ability of the candidate to solve a simple problem they weren't trained to solve.

It is a big problem for the industry. Lacking the ability to put a finer point on problem solving ability.

nomen agentis

But there is also the shortage of what problem

2

Oddthinking

14:09

(The Wikipedia definition is poorly referenced. I offer it as evidence that economists have words for these things.)

DampeS8N

@nomenagentis And even being a programmer that has done hiring doesn't make you an expert in analyzing what is missing from most of the candidates. It comes down to the pornography argument. I can't define it, but I know it when I see it.

nomen agentis

I've also read research showing that a combination of a basic IQ test and work-sample test is the best you can do to assess a candidate

Oddthinking

@nomenagentis While the question referred to Agile programmers, I am sure we could find notable claims for programmers/developers/software engineers.

DampeS8N

It has gotten so bad at most of the places I have done hiring that we actively exclude candidates with degrees because the amount of untrained bad programmers is smaller than the amount of trained bad programmers.

nomen agentis

(That's the obscenity test from Jacobson v Ohio)

@Oddthinking I agree

@DampeS8N the IQ test would get to problem solving ability, and work sample test would get to can you do the job we ostensibly need you for

DampeS8N

14:12

speaking of agile. I have scrum in 3 minutes. BRB. haha

nomen agentis

@Oddthinking do you actually need to find a claim? Or is it enough that people believe it

In this case, finding an example could provide concreteness, but I'm asking about the notability bar

Oddthinking

I come back to the reasons why we ask for a specific claim:
* to make sure it is important enough to bother researching, and not some drunken bar argument.
* to make sure we are clear on what the claim is, and not flailing with unclear definitions.
* to make sure it is a genuine claim, and not someone misunderstanding a joke.

If we are confident about these three things, the actual link isn't important.

nomen agentis

Makes sense

In any case, I agree we could find the same claim about programmers in general

Might be more likely to have been studied. Or wouldn't have to try to break apart populations in surveys

Oddthinking

Talking of problem solving, are we getting closer to a solution here? What is a reasonable question to ask, which is answerable empirically, and gets to the heart of the common claim?

georgechalhoub

I think the claim of "massive shortage of software developers" is clear and valid which has been answered with "there is no lack of domestic graduates to fill available programming jobs".

Many people fell for 'no true Scotsman' by saying some developers aren't true developers.

Can anyone explain to me what isn't clear?

DampeS8N

14:22

back

nomen agentis

Because I think the belief that many people have isn't a raw, unqualified "developer". The version people believe has a quality standard attached

Your answer assumes that the claim doesn't have that, and that is valid, but only for one interpretation of the question. The question could be clarified to be that

georgechalhoub

No.

DampeS8N

@georgechalhoub We are trying to decide if there is a set of qualifiers we can agree on that don't shift and are objective so we can avoid the Scotsman.

nomen agentis

Or no qualifiers so that George's answer works

If the claim really is any developer

georgechalhoub

The OP could ask another question asking if "massive shortage of true, demanding and elite software developers" and find a claim for that.

nomen agentis

14:25

Yes the op could. This question could also be clarified

Both are options

DampeS8N

@georgechalhoub I think the meat of the question presupposes a basic level of competence. Don't you? Otherwise just fill the positions with factory workers or school teachers or the homeless. There already is some set of qualifiers you are working from, graduates. How did you come to the qualifier you chose?

nomen agentis

I wish you all luck! I'm off. I hope you come to a solution. I don't think I have any more insight into this one

DampeS8N

later

@georgechalhoub Your answer basically is saying that the candidates exist but people just aren't hiring them for the open positions. Or are you saying that there are no open positions?

georgechalhoub

It could be both. My answer gave empirical data showing that there are twice as programmers as there are open positions.

DampeS8N

I think it is the former and if the impetus of the question is the assertion of people hiring is that the candidates don't exist, don't you think that warrants some discussion?

georgechalhoub

14:33

It does.

Do you think a person who earned a degree in computer science show no level of competence and obedience?

DampeS8N

@georgechalhoub If they aren't being hired for a position, there is something wrong. Is it certainly a lack of competence? No. It could just be that impression on the people doing the hiring.

But the end result is open jobs without candidates that satisfy whatever standard is being used by those hiring

georgechalhoub

It could be also that they are inexperienced, or their employer require degrees from higher ranked universities.

DampeS8N

If that standard is unreasonable, that's a problem that needs to be be addressed elsewhere. But if the standard is reasonable then by definition the candidate pool is NOT competent.

And your measure is wrong.

georgechalhoub

I also want to point out a big slippery slope you made: "Otherwise just fill the positions with factory workers or school teachers or the homeless.".

DampeS8N

@georgechalhoub It isn't scientific. But I've been in the position to hire for both government contractors and private (video game) industry and in both cases we actively shy away from more educated candidates because they are statistically less likely to be able to solve novel problems.

Trained programmers can solve problems they were trained to solve and with the tools they were trained to solve them with. But for most jobs that is insufficient to succeed.

@georgechalhoub Stipulated.

georgechalhoub

14:40

Aha, sounds reasonable. Then don't call this 'shortage', call it programmers who don't meet quality standards.

In my opinion, if there was truly a shortage, you wouldn't be that selective in picking positions.

DampeS8N

@georgechalhoub It can't be a shortage of programmers that meet quality standards? I see the standard as part of the definition of shortage already. That's were the slippery slope came in. If we can't include a standard of any kind, education is out the window.

@georgechalhoub Bad programmers produce negative productivity. They harm the situation, rather than just work more slowly.

Think of it like this. Yes, anyone can build a wall of stones, but sometimes there are people so clumsy that they knock down all the stuff other people built. Those people get fired because it is better to have the position open than filled with someone negatively productive.

georgechalhoub

I can be a shortage of programmers who meet quality standards. That's why I said the OP can ask this in another question.

It*

DampeS8N

@georgechalhoub This question already lacks a standard to measure with. That's why you imposed your own (education) and other imposed their own, and that's the underlying issue with the question. That's the unclear thing from the start of this conversation.

@georgechalhoub That all was me explaining to you what isn't clear.

georgechalhoub

Okay.

I understand now the confusion.

DampeS8N

:)

Now. When I came to chat today. I was hoping we might be able to find a way to avoid long conversations in chat or comments like ours to get to this. Mainly by having the mods close the question and ask the asker for more clarity.

Does that seem like the right course of action to you as well?

georgechalhoub

14:49

It could be the right course of action, I'm not sure. The OP needs to clarify.

Sklivvz

Closing an answered question is generally not a great idea

DampeS8N

@Sklivvz I don't know the order of events.

Sklivvz

i'm talking about right now :-)

DampeS8N

But I mean as soon as the question was noticed and comments were flying a bid for clarification should have been made before it was too late.

Now is too late. :)

georgechalhoub

I'm afraid if we're going to turn this question changes to "massive shortage of quality software developers", we're turned this into an unfalsifiable question (or extremely broad question) which can only be answered by authoritative statements from Microsoft or Stack Exchange. Moreover, each company has its own quality standards of programmers to hire. Google's quality standards of hiring programmers is different than PornHub's.

Sklivvz

14:52

the question was never that. it was edited into that (by @Oddthinking)

clearly there were (different) problems even before

DampeS8N

@georgechalhoub I agree with you. Which is why this question should have been closed at first pending specific definitions of quality. Or rephrased to address the real concern of there being so many open positions with no candidates.

Sklivvz

right now though, what we can do is try to specify the assumptions in the answers

Oddthinking

@Sklivvz ? I only made two changes - added the USA tag, and closed it.

DampeS8N

@Sklivvz I'm more concerned with what to do with the next question that pops up like this

Sklivvz

@Oddthinking indeed you are correct, i should have checked, sorry :-)

Oddthinking

14:56

@Sklivvz np.

Sklivvz

@DampeS8N I did suggest focusing it on N per candidate

DampeS8N

I have a good bit of experience with questions that have these kinds of deeply embedded poor assumptions from being a protem mod of Science Fiction and Fantasy. You can imagine why. Whenever people bring their own definitions for things to the party, the answers start going wildly in divergent directions. It'll happen again. The solution then was to close it and ask the OP for clarity. I'd imagine it is the same solution here and now. :)

Sklivvz

the OP didn't want to clarify, so it should have been closed. By then though, it was already massively upvoted

DampeS8N

@Sklivvz the definition needs to come from the question, or at least the question needs to state what definitions are possible and answers should address them all.

@Sklivvz I see

That's a hard place to be. I'm sorry.

Sklivvz

the question doesn't need definitions, really. the claim is what it is, it's quite common that we don't have working definitions. However we expect the evidence presented to contain the definitions.

DampeS8N

14:58

The options should at that point be to either edit the question yourself based on what you feel the consensus of definitions are and ask all answers to address all definitions. Or to close the question. Even if it is upvoted highly.

Sklivvz

researchers do tackle these problems as part of their job, and usually do state the assumptions and definitions in the papers

Oddthinking

For the record: Yes, closing an answered question isn't very satisfactory. I see this as an unhappy situation. Yes, closing this question before it was answered would have been better. Finding the line of when to be strict is sometimes tricky. I didn't foresee this one blowing up like this. Part of the reason, I went ahead with closing this one was it was still getting worse and further from a resolution, rather than closer, and I was afraid of more diverging answers.

Sklivvz

I agree @DampeS8N

@Oddthinking exactly.

DampeS8N

It is a tough situation. I know. Imagine trying to untangle it when 8 people have slightly different understandings of what an Elf is. I'm just trying to be helpful and stop something like this from happening again if I can. :)

And to remind @Sklivvz that chat exists and he should bring anything over about 3 comments here next time. ;)

georgechalhoub

A good question would be "Is there a massive shortage of quality software developers at the company X or government Y?" with a claim from the company.

DampeS8N

15:08

@georgechalhoub Better, but it still leaves "quality software developers" undefined. A better question would really look like this: "Are there large numbers of open positions in software development despite there being many more graduates in computer science to fill them?" to verify if the problem exists. The question of why that might be could be a follow up, but I suspect it doesn't really have a clear answer.

georgechalhoub

Yes. I'm also more interested in discussing this question about Manahel Thabet developing a groundbreaking formula to measure distance in space without the use of light. skeptics.se/27566

DampeS8N

It is some combination of there being a natural ability required for being a good programmer, meaning that more graduates and more education doesn't mean there are more good programmers and/or that the training we are giving isn't sufficient and/or that old-gangsta programmers want their wild west while the landscape is increasingly changing to one with ordered solutions to common problems being off-the-shelf and they don't want to give up the control or the pay. It is a messy messy thing.

@georgechalhoub That crap is like a billions pages long. TL;DR

:P

georgechalhoub

One additional point is that a programmer doesn't have to necessarily have a degree.

businessinsider.com/google-hiring-non-graduates-2013-6

DampeS8N

@georgechalhoub I have an AA in game design I got online after bouncing from school to school for 5 years. I went right into working for NASA then on Army.mil and finally made the patcher/downloader for Elder Scrolls Online and this marketing site. elderscrollsonline.com/en-us/esotu Degrees don't mean a lot to me. :)

georgechalhoub

Me neither. I loathe college education (dropped out) and standardized testing. It is an archaic system that used to have a monopoly on the production of knowledge but that has been unable to evolve with contemporary society.

DampeS8N

15:18

Guys who hired me to work on ESO thought the game design degree with cute but it didn't factor into me getting hired. Being able to explain why var this = 105 + "1"; is "1051" in javascript was a much bigger factor.

Oddthinking

@DampeS8N AA?

DampeS8N

@Oddthinking Associates of Arts

the most general and probably worthless 2 year degree you can get

That mine took 5 years is besides the point

That I flunked out of UARTS and Penn College of Technology before finally getting it online from The University of Advancing Technology, one of those Arizona degree mills, is also besides the point.

Oddthinking

@DampeS8N Ah, it isn't a term used in Australia (until 2004, apparently - they were introduced here then), but I know of similar ideas.

DampeS8N

That my father worked for NASA at the time and pestered my first boss to hire me part time. That's much more important. And that I immediately started working full time and got more done than the 4 other people working there while making half as much as one of them is more to the point.

You either can solve the problems or you can't. I've been trying for years to put it in more concrete terms than this with no progress.

I can sum up college like this. In my C++ 101 class at Penn College, the professor wanted us to make a race track application that accepted input like A B C with enter from lists of choices. I ordered the list by most-selected option and let the user use the arrow keys or the letter to jump to the option the wanted and just to hit enter to select.

He didn't know it was possible to do it using the code I provided and it didn't compile in his (different from the class) compiler because it treated Getch() differently. F

Sklivvz

45

Q: Are there more open jobs than available developers?

A recent Stack Overflow blog post claims that: With nearly five open jobs for every available software developer, the need for qualified technical talent is higher than ever. I have seen this claim repeated many times, with different numbers cite, but I've not been able to get down to any u...

united-states economics software

DampeS8N

15:31

Yes, we handed in both the code and the compiled exe. So he knew the application worked.

Sklivvz

I fixed it the best I could

Opinions?

DampeS8N

@Sklivvz Much better. claps

I'd wipe the comments clean

Sklivvz

15:43

@vartec and @georgechalhoub i've edited the question, can you please adapt your answers so they answer the new question? We'll then reopen.

georgechalhoub

@Sklivvz, What do you expect me to add/change to my answer?

Sklivvz

@georgechalhoub Well the question is not about a "shortage" anymore...

ChrisW

oh you mean this chat.

You quoted one sentence, which (in the original) hyperlinks to an article titled Hiring and hirable in 2013: Agile developers which is a much narrower claim than what's being questioned: the current version of the question seems to assume that developers are interchangeable/fungible.

Sklivvz

to be clear, your answer is about the employed, the question isn't

ChrisW

On other sites (e.g. CodeReview.SE) you're not allowed to make edits which invalidate existing questions. So one possibility would be to keep this question closed, and ask it again if there's a better version.

Sklivvz

15:53

@ChrisW the question does mention that the link points to a narrower claim. However the question is an example of a notable claim. The current question does not assume anything I can see, it simply repeats a common claim, hopefully in a neutral manner.

georgechalhoub

To be honest, you have made the answers completely obsolete.

Sklivvz

On Skeptics we don't have such a policy. I would be in favor of it, but not everyone agrees. If we did have that policy though, we'd close first and ask questions later.

georgechalhoub

After both the question and answers have been upvoted, I'd have left the question closed.

ChrisW

"politicians are liars" is a common claim too. Perhaps being a common claim doesn't make it, ipso facto, answerable.

Sklivvz

To be clear, on SE there's no such thing as "leave a question closed" unless it's a dupe. Closed questions will be deleted at some point.

ChrisW

15:56

I thought closed questions were auto-deleted only if they're unanswered.

Sklivvz

It was a bad question we had to close, the answers were also controversial, mostly because of the question. You are advocating we keep that, and not keep a better question with better answers?

This version is clearly answerable. Count the job offers, count the people that are looking for a job, compare the numbers.

ChrisW

It's answerable but stupid. :) E.g. Job ads are for specific subclasses of developer.

nomen agentis

Stupid questions are okay right?

Sklivvz

If relevant, different "subclasses" can be studied

That's part of the answer though

georgechalhoub

Usually, it seems that what they do in skeptics.se is either delete closed questions or lock them as having a significance.

nomen agentis

16:00

In skepticism I've always started broad and then narrowed. I.e. is there something going on, before delving into a more nuanced look or why

So it makes sense to ask the base question of are there more people looking for programmers than there are programmers looking for work, and then later seeing if that is because of a mismatch in desires and qualities

Sklivvz

We can just look at the numbers, then see a split by experience/field and see if it makes a difference. It seems to me like a normal check for confounding factors that would appear in any half decent economics research on the matter @ChrisW.

ChrisW

Is it even a claim, that there are more ads than developers? I guessed a claim is more like, "we have trouble finding (or paying for) enough qualified developers to meet our (immediate) needs".

nomen agentis

That would be a reason for the disparity

Sklivvz

I've heard 3 open positions per candidate many times

nomen agentis

But whether the disparity even exists needs to be established first

If we are following good scientific principles

ChrisW

16:03

people claim to have made various such (specific) observations

nomen agentis

anyway, i think we can definitely answer the question as currently phrased, and move on to more specific questions if there is interest. is there any objection to that? (i.e. why hasn't the question been re-opened?)

georgechalhoub

The viral study which I linked in my answer had concluded and proven that "there is no lack of domestic graduates or existing domestic STEM workers to fill available STEM jobs". In other words, it said that developers outnumber STEM jobs.

And, thus it still answers the question.

nomen agentis

cool.

wait.. developers outnumber all STEM jobs?

Robert Harvey

Wait, what? Aren't developer jobs a subset of STEM jobs?

nomen agentis

that's a huge glut of developers

georgechalhoub

16:12

That's nice sarcasm you got.

that graduates**

nomen agentis

what?

@georgechalhoub Just want to confirm, your answer shows that there are more developers than all STEM jobs?

georgechalhoub

no

My answer proves that STEM workers outnumber STEM jobs.

nomen agentis

Ah, I misunderstood "In other words, it said that developers outnumber STEM jobs." then

georgechalhoub

i meant that graduates* instead of developers

nomen agentis

ah, gotcha

georgechalhoub

16:15

So we got:
1) STEM workers outnumber STEM jobs.
2) Developers are STEM workers.
Can we conclude that developers outnumber jobs?

nomen agentis

but then, that doesn't really answer the question of whether there are more developers in the job market than jobs available for them, unless you also establish that the distribution of seekers/jobs is the same amongst developers as in STEM in general

Can we conclude that developers outnumber jobs? Only if you establish that the ratio of developers/jobs isn't special compared to STEM-workers/STEM-jobs.

georgechalhoub

Hmmm

I'm currently searching the study, they must have included numbers for developers and their jobs.

nomen agentis

Maybe you get my point, but for example, there could be a massive excess of mechanical engineers, but a small shortage of developers, and the STEM numbers overall would still look as they do

georgechalhoub

yeah

agreed.

nomen agentis

I think you're close. You found the right kind of study. Maybe it points to its raw data or something like that.

georgechalhoub

16:26

Yeah. I just found this: "The number of computer science graduates has fallen from its dot-com bubble peak. All the evidence indicates a highly responsive labor market during this period"

Sklivvz

@RobertHarvey the question is fine, the answers though need fixing now. I don't want to have to handle "not an answer" flags ;-)

nomen agentis

what if it turns out that we find that there is no developer shortage

would beyondcoding rethink whether their product is needed?

oh, maybe it's localized shortages

(i'm assuming for the moment that we find that there is no developer shortage... just to think through the implications)

Sklivvz

again, the claim is not about shortage, nor it implies shortages

at least not on a large scale

nomen agentis

what do you mean "again"... this is the first time i've talked to you about this, i think.... okay, i'm using "shortage" as shorthand for "Does the number of open jobs surpass the number of developers "on the market"?"

what if it turns out that we find that "the number of open jobs does not surpass the number of developers on the market". would beyondcoding rethink whether their product is needed?

or maybe it's localized instances where the number of open jobs in that local region surpasses the number of developers on the market for that local region

(i'm assuming for the moment that we find that the number of open jobs does not surpass the number of developers on the market... just to think through the implications)

Sklivvz

16:42

Even if the numbers were roughly equal, there's clearly a lot of friction in the market. Not all available developers are actually people that match the market requirements for example, or have matching salary expectations.

nomen agentis

I agree, but the question is just about numbers disregarding qualifications or other barriers, right?

Sklivvz

So, I'd expect to see shortage-like effects even if there are more developers than jobs

georgechalhoub

Plot twist...

nomen agentis

oh, you're saying that beyondcoding is addressing qualification-related "shortages"

georgechalhoub

The job gap and growth opportunity is in computer science, not in STEM..

nomen agentis

16:43

oooh, in what direction?

georgechalhoub

I seems that comp sci is the only STEM field where there are more jobs than students

nomen agentis

@Sklivvz that makes sense

georgechalhoub

According to some report from the government

I'll update my answer

Sklivvz

@georgechalhoub keep in mind that 42% of developers don't have a CS degree

nomen agentis

even more plot twist

georgechalhoub

16:45

Can you back up your claim?

nomen agentis

yeah, if "42% of developers don't have a CS degree", that makes the numbers from that report less relevant

Sklivvz

yes, look on the stack exchange blog, we did a 2015 survey

it's the largest survey ever done of developers (I suppose)

nomen agentis

eeeh.... i don't like voluntary response surveys. random sampling polls is better

as far as generating numbers that generalize well and are actually representative of a population

Sklivvz

stackoverflow.com/research/…

georgechalhoub

But it seems that the blog questioned 26,000 members only, and they're contributors to StackOverFlow...

nomen agentis

16:48

however, even if 42% is the wrong number, it does show that counting only degree holders is not enough

georgechalhoub

True...

How can we count programmers who don't have any computer science degree?

Seems impossible for me to answer...

nomen agentis

that's part of a good answer :)

Sklivvz

@georgechalhoub sure, but I wouldn't say "only" 26k. That's an incredible number for a survey :-)

nomen agentis

somebody might have tried

georgechalhoub

Yeah, sorry. You're right.

41.8% are self-taught...

stackoverflow.com/research/…

nomen agentis

16:50

add boot-camps, and online courses

those are other non-degree routes to developers.

i think the point is made... that we've gotta find a way to count them too. basically all self-identified "developers" that are looking for jobs... is that less than the number of jobs available

seems like something the U.S. Bureau of Labor Statistics might have looked into

georgechalhoub

It is reported that there are about 18.2 million software developers worldwide

Sklivvz

It's probably much easier to find research done by someone rather than trying to do our own

nomen agentis

yeah, easy route first :)

FYI: bls.gov/oes/current/oes_stru.htm

georgechalhoub

Based on data from the U.S. Bureau of Labor Statistics from 2002, about 612,000 software engineers work in the U.S.

nomen agentis

There are a bunch of codes that fall under what i would consider "developer"

Sklivvz

16:54

Another way to look at this is to look for proxies: how long does it take on average for a developer to find a new job? How long do they "stay on the market"? Compare that to the time it takes for a job to be filled.

nomen agentis

"Computer Programmers", "Software Developers, Applications", "Software Developers, Systems Software", "Web Developers", etc.

Sklivvz

if on average it takes 1 month for a developer to find a job and 6 months to fill a developer job then there must be more jobs than devs

nomen agentis

Sklivvz that would be a different question, but yeah, that's another way to look at it. The proxy approach introduces uncontrolled confounding variables

but what if it takes 1 month for a developer to find a job and 2 weeks to fill a developer job

Sklivvz

then there are more devs than jobs. pigeon-hole principle (i think)

nomen agentis

that seems correct. i thought there would be some missing info, but i can't think of any

Sklivvz

16:58

I think we can't estimate how many

georgechalhoub

That would be original research

nomen agentis

right, but it gives us a one-sided test

two-sided actually

just not magnitude

oh... yeah. we'd be relying on the theory that that proxy is appropriate, which is original research unless somebody else has estblished that

is a "Computer Systems Analysts" a developer/programmer?

Transcript for