fbpx
The Stack Archive

Lack of Stack Overflow or Github posts a ‘red flag’ to IT employers

Thu 30 Jul 2015

stack-overflow-github-interview-screening

The quality of a user’s submissions to development communities Stack Overflow and GitHub can make the difference between acceptance and rejection for some candidates seeking IT or coding positions in the jobs marketplace, according to a new article from Dice.

In the article Willis Johnson, VP at the Robert Half Technology office in Albuquerque, N.M., warns that a lack of positive contributions either to Stack Overflow or GitHub is now practically a requisite for a ‘true programmer’, affirming “If not, it’s a red flag,”

The two communities have a different scope, and are differently characterised in the candidate market by employers. Stack Overflow adopts a Q&A-based format which presents technical questions from users, with ‘Best answers’ chosen from selected responses; in most other respects it differs little in end-use from the VBulletin-style forums which dominated the tech-learning boom of the early noughties. By contrast GitHub is a code-repository used by programmers for versioning, source code management and distributed revision control. SO currently has over 4.5 million registered users, with 9.9 million questions answered. Though GitHub does not publish its user figures so transparently, its own blog noted 3.5 million users when it turned 5 in 2013.

In the Dice article Scott Schefferstein, the senior technical recruiter at New Orleans consulting firm Resolvit, also weighs in on the influence of a user’s posts at SO/GH, refuting all those forum accusations about ‘Grammar Nazis’ by advising members to spell-check their posts and consider the tone of their contributions. “Be aware recruiters are looking. It could come down to whether they like the way you engage with the community. Or, your profile might show that you match a role’s need for subject matter expertise. Either way, it can have an impact on their view of you,”

San Francisco-based IT recruiter Aaron Ho heads straight to SO’s ‘reputation’ score, scrutinizing the ‘Top Posts’ and ‘Top Tags’ found therein. “The assumption,” he says “is that if they’re taking the time to answer specific types of questions, and other people are upvoting them, then they are probably pretty good at it,”

Schefferstein adds that he is able to ascertain useful information about a potential candidate when they make in-depth posts. “Someone who’s glib tells me something else. Read a few of their posts and you can get a sense of them,”

All agree that investigation of a user’s activity on either site cannot be used as a de facto index of a candidate’s suitability for a position. Schefferstein notes “Some of the best developers I know, you wouldn’t be impressed by their Stack Overflow profiles at all,”, whilst Ho observes that frequent or even apparently assiduous or talented contributors aren’t automatically candidate material: “You have to take a look at the content of the answers and how well they were received by the community,”

Though Stack Overflow spun out as a ‘popular’ iteration of the less chaotic Experts Exchange in 2008, its apparently organic ‘crowd-sourcing’ of tech wisdom belies its rather more formal relationship to employers and head-hunters. SO has an employment portal relating to its user-base, whilst GitHub launched its own ‘Jobs’ section in 2009.

Best behaviour needed in ‘casual’ online environments

Any regular user of Stack Overflow will know that it’s an informal environment where a rolling squadron of moderators squash – or at least attempt to impede – irrelevant or duplicate questions, flame wars and all the usual kinds of inadvisable behaviour which has made Facebook such a controversial place for employers to seek information about potential employees. The ambience of free expression in its posts is largely in opposition to the formal and business-like material we might consider submitting to LinkedIn – which is probably precisely the reason why head-hunters and potential employers view it as a chance to see the programmer as he or she ‘really is’.

When a user at parent site Stack Exchange asked ‘Will high reputation in Stack Overflow help to get a good job?’, the ‘winning’ answer responded ‘Having a high reputation can’t hurt, but it isn’t a magic bullet,’ and added ‘Are these people good programers? [SIC] Undoubtedly yes! Does that mean they are a good fit for your team? Absolutely not. Calling these people “superstars” may be completely correct, but that doesn’t make them perfect,’

Other responses to the question observed: ‘My line of thought is that if you can score 10000+ points by giving meaningful answers, you probably have a considerable amount of knowledge about some technology,’; ‘I hope it doesn’t. I’ve seen some really stupid questions asked repeatedly by people with high reputations and stupid answers by others,’; and user Mike DeSimone additionally noted that many people accrue high reputation by being quick to respond rather than necessarily providing a high quality of response.

Unusually this particular question was given an alternative and opposing ‘best answer’, where Stack Exchange co-founder and CEO Joel Spolsky said:

“Come on, seriously: look at the first page or two of Stack Overflow users. Pick anyone at random. Look at three or four of the highly voted answers they wrote. If you’ve ever hired a programmer in your life, it’s obvious those people are all some of the best programmers you could ever hire.

“Then keep going deeper and deeper. Scroll to page 5. Edit the URL and go right to page 100 where they have reputations in the 3000 range. Look at everyone. With the very rare exception of someone who got a lot of points for a silly answer, these are all obvious superstar programmers… the kind that most teams would kill for.

“Will this actually get them jobs? Ultimately, we’re betting it will. You’ll have to find a company where there are actual technical people making hiring decisions, not resume-reading-monkey-recruiters. And you may have to find a city to live in where there are a lot of good programming jobs: if you really, really don’t want to leave Roswell, New Mexico, you’ll be limited to a vanishingly small number of telecommuting jobs. But if you’re in the top, say, 5000 Stack Overflow users, and you can work in Silicon Valley, New York, Seattle, Boston, Austin, Research Triangle, London, Hyderabad, Bangalore, Sydney, Beijing, or Tel Aviv, I would be extremely surprised if you aren’t in high demand and earning in the top 10 percentile,”

The legality of researching job candidates based on their user-activity

Though it’s a subject that isn’t limited to tech forums or coding and IT jobs, employers are obliged to consider their legal obligations when employing web searches and user-posted information to furnish information on the suitability of candidates. A post at Compliance and Ethics points out the hazards of seeking out web-based material which would be suitable candidate-assessment criteria and inadvertently unearthing potentially prejudicial facts or indications about the person in question. In this regard the article cites the 2011 case of astronomer Martin Gaskell, who won a settlement of $125,000 from the University of Kentucky after their online researches about his candidacy disclosed his Christian beliefs, which led to a decision not to hire him for a top post at the university’s observatory.

Candidates who feel aggrieved that there may be nowhere for them to truly ‘relax’ on the internet shouldn’t be too encouraged at such legal outcomes, however, since cases leading to an investigation of an employer’s email chains or internet browsing history are infrequent, and in any case will be unproductive if those undertaking the research make use of a VPN while they’re doing it. The message seems to continue to be: if you’re going to unwind, then unplug too.

Tags:

employment news
Send us a correction about this article Send us a news tip


Sorry. No data so far.


Do NOT follow this link or you will be banned from the site!