Walmart Labs: Why we run an open source program | Hacker News Reader

Walmart Labs: Why we run an open source program

todogroup.org

163 pointsjamesgpearce68 comments

The community is having a great discussion around “remote work”. You see the dichotomy of companies such as Slack who build products that happen to enable better remote communication, but who hire people to co-located buildings.

As soon as you get beyond a few people, you are working “remotely”. If you aren’t in the same room you will have your main workflow happening through tooling. Yes, you can get together to meet face to face on topics, but that isn’t your general workflow.

Never heard such glory come from anything associated with Walmart before.

Remote working is what you do everyday even if you are 'co-located in the same building', or same company different building. To top it off all your customers are remote, get your communication and systems in check to also do remote, it gives an external view like no other on your product, and gives all your product engineers a lab to work in.

I was having a conversation last month about Walmart Labs. apparently they try very hard to point out they are NOT Walmart. :)

Somehow I don't find that surprising at all.

http://www.forbes.com/sites/clareoconnor/2014/04/15/report-w...

Apparently they have a hard time recruiting as well due to the name.

Thats understandable. Walmarts brand is pretty much defined as "We treat our employees poorly, but look how cheap things are."

I wouldn't consider either of those factors good selling points for an employer.

I work for Walmart. Remotely. It is glorious. While many of the teams are Scala and clojure focused, my team does mostly Golang.

Walmart seems like a great place to develop for. They use some awesome and fun technology.

How could I get in touch with a team within Walmart that uses Scala? (To see if there are any available positions.)

Here's one job post mentioning scala: http://jobs.walmart.com/silicon-valley/engineering/jobid6945...

walmart.ca is now running play and scala afaik: http://www.nurun.com/en/news/nurun-launches-redesigned-trans...

How do I know what teams work with Clojure? The jobs site seems to just mention Java.

Staff Software Engineer: http://jobs.walmart.com/silicon-valley/engineering/jobid7058...

Google seems to be the best way to search for these positions.

The majority of great engineers have GitHub profiles and have been working on projects there (open source and otherwise). The stack has become ubiquitous, and the majority of engineers either like it, or are “happy” enough with it. There are a few grey beards that rant on about Perforce or something else… but that is pretty rare :)

The majority of great engineers... in what field? Web development? Probably. Some language communities are also disproportionately represented. As a whole though, I can't buy that everyone has been totally assimilated into GitHub.

I also like how everything other than GitHub (not even Git) and Perforce has just about spontaneously evaporated according to this paragraph.

Classic selection and confirmation bias.

I have a GitHub profile > I'm great > I only hire great people > I only hire people with a GitHub profile > Everyone I hire is great and has a GitHub profile > Everyone who is great has a GitHub profile.

Agreed. You need to recognize that there are some of the best engineers in the world work at companies where they aren't able to put code on GitHub, or it isn't worth it for them. You're working on something secretive, you have a family and want to spend your free time with then rather than weekends dealing with pull requests, etc.

This mindset also creates and incentive for employers to keep their engineers from working on open source as much as possible, so that they don't look attractive to other employers and are never able to jump ship / ask a pay raise.

In general, I think it is a very lazy and ineffective answer to the "interviewing is hard" problem.

I must be in the minority. I am the greatest engineer in all fields, and I don't use github because my job is not github friendly. I only say I'm the greatest because, hey, if we are throwing around value judgements about engineers I'm going to grab me some. Actually, I wish that practice would stop. Every blogger seems to know who is the greatest, and not surprisingly it's always themselves and people they know using the same technology as they are.

Personally, I'd rather have employers judge me on code I've written for projects in my GitHub profile, code that I've meticulously architected and refined over periods of weeks, months and years, targeted to solve non-trivial, relevant real-world problems in an elegant and expressive manner, over some code that I'm forced to spit out on the spot under unrealistic time constraints solving nerdy CS algorithm puzzles and doing low-level performance optimizations that's never realistically worthwhile to implement from scratch in production (at least for the kinds of problem domains I'm interested in).

I do understand the possible misrepresentation resulting from judging candidates based SOLELY on GitHub profiles, but for those with presentable contributions/projects readily available on GitHub, I feel it makes for a much fairer representation of the candidate than the alternative.

Me too, and that's why I set up our hiring process such that candidates are welcomed to supply code samples via github and other means if they can. However we recognize that this simply isn't feasible for everyone, and it's not held against anyone if that's how it is.

For me personally, nearly all of the work I've done over the last handful of years is on github, however most of that is contained in private repos that I can't share. I don't think that's what people mean when they say that I'm supposed to have everything on github :)

"...(open source and otherwise)." - how will that "otherwise" help?

I'm not going to comment on who is writing the article because that's not the point.

I think this is a good article pushing for more open source development at traditionally large enterprise software companies. It helps the open source community and it helps the company attract talent. It also gives good advice to large enterprises saying that they shouldn't just throw their code into open source without good documentation and community engagement. At the very least its a good start for companies that would have otherwise kept all their code internally.

Just because the article mentions Walmart Labs working on hapi (apparently an express-like nodejs framework) doesn't mean that HN should suddenly disregard the content of the article (as some of the other comments seem to be doing).

Good points. This can be a good article for people in large companies who need to justify Open Source efforts.

Yep I plan to post this to my company's intranet social network. My understanding is that our company's policies prohibit contributing to the open source projects we use.

im afraid this article will fall on deaf ears to those folks. most c-level folks are all about protecting intellectual property and racking up patents. indeed, there are special little ceremonies for patent awards, along with bonus at most enterprise corps.

If you have 40 minutes to waste, this talk from a Facebook manager talks about this problem: https://www.youtube.com/watch?v=fzL6Zoy_ndk

"The supply doesn’t meet demand, so it is important to do everything you can to help attract, and retrain talent."

Here we go again. Please stop lying! Property supply also doesn't meet demand, guess what happens to prices? Salaries are static => no major demand.

"The supply [of 10x+ programmers] doesn’t meet demand, so it is important to do everything you can to help attract, and retrain talent."

People who say this kind of thing won't look twice at someone with a bachelor in Java and C++.

Price elasticity is also a factor. Increasing demand does not guarantee increasing ability to pay.

Dion's a smart guy and a great communicator and I'm really excited that he's been pushing hard at Walmart Labs since moving on from Mozilla (and Palm.)

Walmart does things at scale and if they can bring some of that strength to open source, even just improving business understanding of open source (ideally a lot more, as Dion expians) that's nothing but solid good for the world.

Walmart Labs talks about wanting to bring in great engineers, and yet they make prospective engineers piss in a cup.

Glad they've got their priorities sorted out.

Is this true? Walmart Labs was sounding pretty good to me after reading the rest of this thread, but I would never work for a company that does drug testing. If anyone has the true story here I'd be interested to hear it.

Walmart do say on their hiring page that they require a drug test unless prohibited by state law[1]. I would assume it applies to their subsidiaries as well. It seems rather unnecessary. I can sort of understand it for jobs that require security clearance or where there's a risk factor (a friend of mine had to do one when he took a tech job at a bank). I don't quite see Walmart fitting into that category though.

[1]: https://hiringcenter.walmartstores.com/OnlineHiringCenter/di...

Yep. My roommate got an offer from them last year. That offer was contingent upon a drug test. He ended up going with another offer.

I can confirm they do drug tests.

> but I would never work for a company that does drug testing.

I wouldn't drug test employees, but I can understand why some companies might be compelled to do so. I have friends who smoke pot recreationally, and they're fine. I have also had friends/coworkers who became genuinely addicted to drugs, and very consistently they ended up as ticking time bombs for their employers. Things didn't end poorly, they ended disastrously. I know that's not always the case, and that it doesn't justify drug screening everyone, and that it's an unpopular opinion here. But I've seen enough that I know why some employers think it's worth it.

But you can use existing workplace protocols to deal with the drug users who are bad employees while also ignoring the drug users who are good employees.

I can also confirm that WalmartLabs does do drug tests after accepting the offer and before orientation. The fine print says they can also drug test you at any time after starting employment, however, I have yet to see this happen in practice after three years.

I can confirm it was a request made when I was contacted regarding a position. I lost all interest at that point.

It's not about drugs (though that really shouldn't be a concern). It's about respect, privacy, and power relations.

There are companies which have mandated drug-testing policies by regulation (mostly involving transportation or pharmaceuticals). It's still a major chilling factor so far as I'm concerned.

I would gladly work at one. I have worked around people who did drugs with all the claims of how they were responsible. I will never subject myself to that again.

I did the piss test prior to my current employment and haven't in nearly twenty years since then. About the only chance it will happen is if you give clear signals your messed up.

Plus many industries are required by law to test. Considering the breadth of WalMart operations its likely they cross one of those laws and to avoid discrimination they test everyone.

The problem with drugs tests is the test itself. Drug tests are very effective at finding thc metabolites (30+ days in some accounts) but rather bad at finding those for heroin, cocaine, ecstasy etc (1 - 4 days). So in reality its less of a 'drug test' and more of a pot test.

"One quick anecdote: a great engineer that I know well was recruited to work for a top class company. They basically lost him when he was told that he would be working with an old Java stack and his workflow would not be git based. The tools matter."

How prevalent is this for "great engineer" looking for work? I guess that this type of criteria would only be restricted for those working in the web based. Other industries, not so much?

For me, a great engineer is someone who can create a great product with whatever tools are available for them to use.

Yes... but if you're a good engineer, why would you work with shitty tools? All things being equal, why would you choose to work with tools you find frustrating when you can work with tools you like?

That's the point: tools matter, and by holding on to crappy tools, you might be giving people a reason to not work with you.

Because, sometimes, shitty tools is all you have to work with. I work in embedded systems, and some of the tools I have available is absolutely the shittiest. Imagine debugging a realtime OS on a 8bit micro by flashing an error code through a LED.

But is that because you don't have any other options available, or because your company told you to use inferior tools?

This was the only option. The other debugging options were too intrusive and would colour the information.

Ok, but we're talking about a website, where the toolset is limited by political decisions, not technical ones.

Maybe they can create a great product with whatever tools, but do they want to? Don't many experts want to choose among the most appropriate tools? (Not to mention other facets of the working environment they spend half their lives in?)

Great engineer is relative. I'm a "great engineer" and I will work with BASIC or PERL and CVS if it came down to it. The problem I solve is more important than the tools.

I think the point is that great engineers can often be choosers as there is no shortage of jobs available to them. Given the choice between a job with shitty tools and one with great tools, he/she would choose the latter. Ditto when it comes to company culture too.

" but we could bring in a world class team of engineers who were desperate to build very large scale node services."

The word "desperate" is interesting. Are engineers really desperate to build "very large scale node services" (or whatever)?

1. Yes but only if technology 'X' is in-demand 2. Yes but only for those who like to try new technology and gets paid having fun :)

Am I the only person that read through a good portion of that and didn't actually come away understanding "Why (walmart labs) runs an open source program"?

It looks from their website like they have some really interesting work that they've published, but I'm not sure what the utility in this sort of PR fluff piece is.

There's an interesting balance between maintaining a competitive edge and helping advance the field as a whole. Overall I hope companies lean towards the open-source mindset. Working on problems that someone else has solved is a lot less fun than pushing the frontier.

a group of amazing engineers, unfortunately, many are leaving

Why?

Are Walmart Labs engineers locked in the store overnight and forced to work off the clock?

You know what's cooler than working for Walmart? Working on disrupting Walmart, and forcing them to increase the quality of the products they sell to keep competing.

Huh? They're competing just fine by keeping prices down, which is what most consumers look for first.

You'd have to disrupt our current socioeconomic situation.

So unless you can beat their prices and their quality, good luck.

I do that by going to Costco much as possible.

Costco has something like 1% of the product selection that Walmart has.. let alone online purchases... Given, I tend to order a lot via Amazon, but no brick and mortar store comes close to the U.S. market penetration that Walmart has. Most of the population (mid-high 90's iirc) is within 5 miles of a Walmart.

Say what you will, but they're probably about the only shopping choice in competition with the likes of Amazon.

http://www.bookdepository.com sells books for cheaper for free shipping to Australia and is much better than Amazon for books IMHO.

The book depository is owned by Amazon

Which is fine, but I didn't mention books.

> Huh? They're competing just fine by keeping prices down, which is what most consumers look for first

I dispute that. If consumers looked to price first everyone reading this would be using a ChromeBook, but I imagine a sizable number are using MacBooks instead. A more generic example: some consumers will buy the cheapest possible clothing, but most consumers look to a reasonable compromise of price and quality.

This is an obtuse simplification. I would be happy with a Chromebook if I didn't happen to be a developer. Also, it's not easy for the layperson to measure the quality of groceries. A lettuce is a lettuce, so I'll buy it where it's cheaper.

Just because you're price sensitive, that doesn't mean that you'll buy something you don't want, just because it's cheaper than what you do want.

There's been plenty of research that shows most people are more interested in lower prices than making some sort of social statement.

True. I think most people want the lowest price for a specific branded product. They start off by deciding, more or less, on getting a certain level of perceived quality, and then finding low prices. The price is secondary to making the decision to purchase a product.

As far as making a "social statement" - buying most recognizable or branded products is a kind of social statement. If it weren't, we wouldn't have such a diverse selection of cars, clothes and laptops.

That's perhaps one of the worst examples you could give to support your argument. The vast majority of people are not developers or otherwise employed in well compensated IT jobs. They most certainly look to price first when it comes to personal computers.

Disrupt Wal-mart? Best laugh I had all day.

Reply on news.ycombinator.com