"MongoDB ... takes the lead for the most wanted database."
I'm not sure that's a title I'd want.
Stack Overflow’s annual survey has revealed the tools and tech that developers love to hate: Visual Basic 6, IBM Db2 and SharePoint. According to the poll, which took in the views of more than 100,000 devs, Rust is the most loved programming language for the third year running. It is closely followed by Kotlin, which makes its …
"the majority appear to be straight white men"
That sounds to be pretty representative of software developers. Normally, a poll which is intended to discover what products software developers are using ought to be polling a representative sample of software developers, not a representative sample of for example trendy social media advertising consultants or PR flacks.
If there ought to be any concern about how representative the sample is, the concern ought to be with respect to how well the sort of person who answers Stackoverflow surveys is representative of the sort of experienced and knowledgeable software developers whose opinion on matters of what software is good or bad is worth listening to.
'More than half of respondents said they wouldn't write code for a product or purpose that was “clearly unethical”'
I used to work on missile guidance systems. I was never comfortable with it, but those missiles have never been fired in anger and have never hurt anyone.
My current employer is one most would consider much more ethical, but they've contributed to preventing the treatment of hundreds of sick children and adults this year already.
I'm considering moving into Internet gambling, at least that way you know how unethical you're being.
There's a stupendous amount of legacy software written in VB6 so they need to support the runtimes for as long as feasible. That does not mean they are encouraging developers to use VB6 for new programs.
There will come a time when Windows will no longer be able to fully support VB6, probably in 3 or 4 versions down the line, that'll be interesting to watch.
VB6 support will only disappear the day Microsoft is no longer present in banks and Fortune 1000 companies.
The year 2018 is still young, but I am sure that, just like in 2017, I will be called upon to teach bank personnel how to code in VBA.
You guess how many years that is going to have an impact.
VB6 is the new COBOL. It will exist as long as the human race does
Yes. There are some languages - COBOL, VB6, Java, Excel - which for a combination of reasons reached a stage of deployment in business IT where they will continue to be used for a very long time yet. The economics simply do not support replacing them.
Fortran is similarly entrenched in scientific computing. While R and Python (and some lesser contenders such as Julia, and again the accursed Excel) are widely used, there's a large legacy base that will last for quite a while yet.
perhaps longer if sentience happens.
This, on the other hand, is mystifying. Sentience has already happened, and continues to happen, all over the place. It's one of the few logically provable questions in the philosophy of mind, thanks to that Descartes guy. Perhaps you mean "the singularity", in the Kurzweilian sense?
 Yes, Excel - however horrible - is a programming language.
 Under certain assumptions blah blah blah.
RE "This, on the other hand, is mystifying. Sentience has already happened, and continues to happen, all over the place. It's one of the few logically provable questions in the philosophy of mind, thanks to that Descartes guy. Perhaps you mean "the singularity", in the Kurzweilian sense?"
"Perhaps longer if sentience happens" was a lapse into fortunate whimsy, to which I am prone. I imagined the vast VB6 project I am in process of converting coming to me in the night as I sleep and snuffing the one who is trying to terminate it while it still can. Coding requires whimsy. I certainly do. Not sure I agree with the singularity...maybe...not for a while...
"There will come a time when Windows will no longer be able to fully support VB6, probably in 3 or 4 versions down the line, that'll be interesting to watch."
But, but, but.... Microsoft has already stated that Windows 10 is the last version... ever (until it's not ;) )
So what you're really saying is VB6 will be supported forever. Long live VB6!!! (not).
The poll was of developers, the fact that the majority of those answering the poll were straight white men means the poll was aimed at the target audience as the vast majority of developers are straight white men.
Diversity should never be a priority, the top priority should be hiring the best talent irrespective of who they happen to be. It's not the fault of employers that the majority of those who studied for development roles are straight white men, there is nothing stopping anyone else from learning how to develop software, they simply chose not to.
Yes, this is a good question. Being a straight white guy myself I can only give some anecdotal evidence from tutoring students (school to university level) and chatting to my fellow students:
- There are still too many people being proud of not understanding math (or even basic simple calculations)
- There are still people saying that it doesn't matter if you are not good in maths and sciences if you are a girl - some of these people are female (many, actually) and some are in the above category. I hate that!
These two are probably the main culprits, because you need to get people interests in that stuff early on!
Addotionally, they tell me that being in the male dominated STEM subjects can really be weird for a non-male. Doesn't even have to be harassment, but not being one of the boys is difficult. Too many try too hard to be one of the boys, some end up as s...s (sorry, my observation). Being yourself in that environment is tough. Makes sense to me, and it's hard to overcome this culture. I hope we are improving...
My observation is that the gals who studied with me had made a more balanced decision than me (and most of the other guys). Thus they were (on average) better than the guys. None of them dropped out (though one changed to maths and one to chemistry - close enough that it is not "dropping out" to me). I think none of them still is in science - they left for better paid jobs.
Don't know how its in UK universities but elsewhere that has somewhat changed already.
That being said it's going to take time before it actually impacts hiring.
If you are looking some senior +5 years in the belt having more diversity in 3 rd year students is not going to make a dent until 8 years time...
Math too hard, too nerdy, too many males already, too much work, low pay, not 'cool'.
Humans are herd animals and that's painfully obvious to anyone who have a brain: you study the same thing your friends are studying. Or parents.
That means artificial 'diversity' is a goal which doesn't make any sense at all: You choose competent people and who cares what they look like. Only incompetent people are worried about that.
Choosing people just by the looks or gender means competence is irrelevant.
That's just a 'good fellow' -club again, just different people favored. It led to major incompetence at every level, why anyone would believe that doesn't happen again?
"Choosing people just by the looks or gender means competence is irrelevant."
That's how most IT hiring seems to go as far as I can tell. My talent and competence has always been over ruled by my long finger nails, odd hair, lack of a business suit, etc. Doesn't even help that I'm a straight white male. On the other hand, I can't recall the issue of my sexual tastes ever coming up, they probably just assume that this white male is straight.
> So the question is: Why do so few non-white males study software engineering in the first place?
Because it's not cool. Computers are cool, and have been since the 90s, but the career track isn't.
Specifically, if they want a high-paying job in the field: A) Their friends and families may shun them for selling out. Some respect "making bank" but many don't. B) They'll be diversity tokens surrounded by whites/asians/gays/SJWs - NOT COOL. C) They'll be expendable cogs serving an Orwellian fascist police state, ultimately oppressing their own people - NOT COOL.
Whereas us white guys only need to worry about creating a police state where all subjects are equally poor and oppressed. But sadly most white people don't give a fuck as long as you're making bank and working on trendy tech shit.
there is nothing stopping anyone else from learning how to develop software, they simply chose not to.
If you don't understand the shallowness and falsity of this claim, you are part of the problem.
There's been plenty of research on this issue. Do yourself a favour and look it up.
"If you don't understand the shallowness and falsity of this claim, you are part of the problem."
Nope, that's a literal truth. If you refuse to admit it, then you are the problem who is barking at the wrong tree and you haven't even read the research. Or understood what it says.
"Microsoft’s research found that girls in the UK become interested in STEM subjects just before the age of 11 but this drops sharply when they turn 16. In addition, less than half (43%) of those surveyed said they would consider a career in STEM."
So most girls don't even consider the sector already at school and you say that there are not many females working in the area because they are not hired (because of gender).
Which, of course, is patentably false, the selection happens a long time before college/university and even longer time before these people are actually applying for job.
You could blame schooling system for that but that's something you don't do. Why?
Anyone have actual numbers? In my software division of (well-known megaco) dev numbers are about 3x in Asia what they are in the US and Europe. Of the US employees (mostly Cali) we're something on the order of 20% female, 70% Asian descent, 5% Hispanic, 5% African or African-American. Of the white males perhaps 10% are gay. I'm not saying it is a diverse team, but it's not 90+% straight white male for sure.
Is this just a California thing? We were an acquired startup a few years ago, so maybe it's that? I'd love to know real stats for developers, not just companies.
"Diversity should never be a priority"
it is not a "priority", lack of it is idicative of a problem.
If 90% of devs are men, it impies stongly that there is bias in the system.
This raises the concern that intelligent female minds are not taking up the the profession and stupid white men are filling roles that could be filled by people who grok stats.
A valid concern... come to think of it, I can't think of a single incompetent female developer I've worked with, and the good ones represent about half the good developers I've worked with. Could it be that most women who are cut out for the profession are, in fact, driven to enter it? And those who aren't, tend to enter female-dominated low-skilled professions instead?
Perhaps there is no problem. There are few jobs for unskilled men, and 90% of developer jobs require no skills whatsoever.
On a copy-pasta site aren't really indicative of anything other than the number of people using copy-paste because they didn't know how already. So either the language is poorly designed or for some reason attracts the opportunistic who don't want to just buy the book(s) and practice.
That's not much of a recommendation...
Follow the money.
It doesn't look so, if half of those who answered the poll develop on Windows.... it's very strange they think Linux is so cool, and then don't use it. A large number use Visual Studio as well.
So it looks this poll returns more what developer thinks they should use to look cool, instead of what they actually do.
Windows and Visual Studio usually are what you get form your employer, not necessarily what you actually want to work with.
It's a compromise, if it was for IT departments all developers would have to use VBA so they can have a consistent image for all clients.
"it's very strange they think Linux is so cool, and then don't use it."
You know ... .developers do whatever they are paid to do? With tools they are given, mostly.
Linux-developers are a minority and many of those don't need SO, so not participated either.
Not strange at all.
Most dev is _for_ Linux. All super computers are linux, most small embeded os computers are linux, most servers are linux, cloud is generally linux. Windows still has a large share of desktops. Therefor lots of people use Windows desktops to develop code for Linux. Seems reasonable.
Trying to persuade my IT dept to let me run desktop Linux when I only ever code for Linux has got nowhere in many years.
Not surprised to hear that this is an industry trend.
Most of the languages mentioned are very, very niche and wll be gone tomorrow as quickly as they arrved here today. No solid OO languages or C. Stuff where code is hacked out rathe than the means to express a design (which is why they get so tied up in language wars) . And if all you need is a quick and easy DB, use MongoDB if you want, but it ain't replacing DB2 and Oracle anytme ever in large companies with very large database requirements, who need something as robust as possble with huge processing capabilities. That's why they're "difficult" : because that stuff ain't easy.
Trebles all round...
I don't undestand this hate for Oracle as a tool. There's a lot to hate about Oracle as a company, but Oracle as a database is something that is far more powerful than most of its competitors, albeit it is so vast it requires a not small effort to learn how to use it effectively, and the size of its documentation is daunting.
There's very little you can't do in SQL or PL/SQL, and often you can solve complex needs without ugly hacks. Sure, if you use it as a simple data dump it's horribly complex - but if you do that, you didn't understand what a RDBMS is - hint - the key is the letter "M".
Also most used language is meaningless.
It's probably similar for many people, hence even though 99% of my work is not JS, JS would feature on my languages used (as would perl - little used, but language of choice for short notice bulk edits of large text data e.g. CSV files as quick to knock up a script and very performant at text processing)
Apply similar logic to other people and JS is "popular" in terms of JS has been used by lots of devs
Caveat: I dd not fill in the survey
.. Also do imagine its quite likely to disproportionately attract JS wielding "hobbyists" as mentioned
So you are saying people who build highly successful web/mobile applications aren't professional developers? If you make a living on something is that not your profession?
That being said there are a lot of people without a clue or with limited industry experience going round StackOverflow but that's fine. It's a place to ask questions after all.
Python is an important language. It's dynamic and therefore limited when used with large teams, but it has really great support on data science and machine learning side. Furthermore, for a small team, it's a great way to get your web app or API up quickly while building something that is also secure and maintainable. Just check Django's out of the box security & project structure. Beats most ways of building same things with JVM languages with the exception of modern versions of Spring. Somewhat surprisingly writing Python also pays better than Java for the same job.
That said, I agree with that if you have large long live applications they should be written with a compiled language, simply because tooling support where the project is too large to understand for anyone developer entirely. Does that have to be Java? Not so sure.
Big things are being built with Go as we speak cool opensource examples being Docker, Kubernetes, Ethereum, Terraform, and Vault. For-profit world besides Google, Github, IBM, AWS, Microsoft and Pivotal some financial institutions and most Chinese mega Corps are picking the language up as less verbose/complex alternative to enterprise Java. Here is a good bit what the language is about https://www.youtube.com/watch?v=k9Zbuuo51go
The whole thing is to build for maintainability and software engineering at large scale. Not something Java was ever purposely designed for. Will it make sense to migrate some old core banking stack to it? Probably not. But it will be a really good tool to build new better one. Look at Monzo https://monzo.com/ in the UK they did it.
All that rant aside I don't hate Java(wrote some yesterday) I'm just saying keep up with new stuff to keep your options open rather than fortifying on the old tech stack. All it takes is a new company CTO/CIO and your job could be gone with that mentality.
VB in Oz has a better known meaning - A mass market beer >>====> This fuelled many a Friday "work" afternoon when VB6 was new and shiny, helping to numb the pain of those who were transitioning from VB4/5. If you want staff to learn to "like" VB6, get them to do a couple of projects with VBScript.
"Git is the dominant choice for version control, with 87 per cent saying they check in their code via Git. Some 8 per cent said they copy and paste files to network shares, another 8 per cent said they zip file back-ups and 5 per cent said they don’t use version control."
That's 108% of respondents accounted for already and we haven't added in figure for SVN, Hg or whatever Microsoft's default Team-thingy is these days. But then, this is a SO survey. I'm surprised that 4% didn't reply "where's my chocolate ... you promised chocolate ..." to every question.
"Microsoft's default Team-thingy is these days."
Git in VSTS and I believe Git is also encouraged in TFS too. "Teams" is an office product and offers no source control but can host a tab showing your VSTS/TFS site which itself houses a Git repo.
Also still offers the MS version control for those who want it.
Developers dread SharePoint, no wonder! Stoneage tech: dotNet 1.1 WebForms - 2003 is calling.
These are just web apps are using additionally on of the SharePoint APIs. They just have a themed back-button and have otherwise nothing to do with SharePoint at all. The SharePoint beast, of dotNet WebForms decades old XHTML code spitting monster, with tables as layout and iframes to include new "apps". Anyway storing data in SharePoint means storing data in flat lists (similar to excel spread sheets). A far cry from real relational databases. And yes InfoPath is dead, but no successor in sight. So digitization of traditional business forms has no clear road forward at all. And revision control in SharePoint is a shit show. And workflows is a big unstable mess. And the joke solutions from the big two third party providers. Especially K2 forms and workflows are a bad joke. Anyone mentioned SharePoint designer yet? The decades old Frontpage app renamed "SharePoint Designer" but lost of its features over the years. Better stay very far away from the mess that SharePoint is.
Yeah but you are out of date. Hate all you want but know what you are hating.
They are not just web apps, they live and are deployed into the SharePoint system and pages and are supported by the SharePoint framework. WebParts are built using React. That is different to having a separate website/app.
hit an Office 365 SharePoint site at _layouts/15/sharepoint.aspx and its fully responsive. There are updated version of the lists and customisations are done clientside. What it runs behind the scenes you don't need to care about.
What document management system have you found that is better? Google Docs, Alfresco, OpenText?
It's is a solid document management and search engine that integrates nicely with Office.
Its easy to take the bits that are useful and ignore the rest.
You could always look at free software like Alfresco, i can put you in touch with people who use it daily and hate it with a passion.
Well DOS screens were extremely fast for data entry. Users hated Windows for data entry. I'm not going back there.
You can create crap UI's in any tool, some people have a talent for it.
Sorry about your day, on the bright side its almost Friday.
"What it runs behind the scenes you don't need to care about."
As a user I see whole document tree structure in every URL Sharepoint shows me and those are 3 lines long when copypasted into a document as links.
Not only that, it _still_ has unique hex ID to refer to a document in that one directory but if you omit the directory tree it patently doesn't find the document, dispite having unique (and very, very long) ID.
That's a major fail in design level.
Also it's very, very slow: Finding a document with an ID should take milliseconds, not tens of seconds, regardless the amount of documents.
There is only really three ways to refer to web documents regardless of tech, Contents (hash), Location, ID.
SharePoint does not do Content (aka CAS) and for often changing documents that's not a great loss but does do Location and ID.
You can switch the Document ID service on and all documents can be linked via the DocIdRedir.aspx page, no multi line links needed.
If your system is spending 10 of secs to find a document you probably need someone who knows what they are doing to take a look. Document Id and links can be found via a sub-second search in 100's of millions of documents.
If you are putting a large number of documents into one library that's a design flaw. Simple steps can avoid problems.
This just in:
Mainframes are now extinct as polls in the 1990s said they would be.
Cobol is completely out of DP as polls in the 1990s said it would be.
All enterprise IT is done in browsers running java applets as polls in the 1990s said they would.
And my favorites du jour:
England now 350 million quid per week richer thanks to a poll a couple of years ago.
The United States of America's political swamp is now drained thanks to a poll last year.
Biting the hand that feeds IT © 1998–2019