345 karma · joined October 16, 2014
Article A: Here’s a photo of a another galaxy so far away, showing XYZ.
Article B: We suspect there’s another planet in our own solar system but we can’t find it and supposedly it’s too far away for us to photograph it.
As an employer the ideal is to pay at or slightly above the market. Enough to make money a non-issue, but not so much that people who secretly hate you still work for you.
Except it wasn’t “All Users in your AWS account”, it literally meant ALL USERS of AWS. Disaster. This led to the same problems as this Jira issue.
AWS changed it so that’s no longer an option. Atlassian should do the same.
You can fulfil the order and make $XM, or you can report it and get... nothing.
Now as the employee of course you want to do the right thing, so you run it up the chain. And someone says to the board “we should invest millions in better systems and people to detect suspicious orders, so that we can make less money but do the right thing”. Any wonder this didn’t happen?
The blame here seems to be with the design of the regulation/oversight. Where was the FDA/other agency reviewing a random sample of those orders and then smashing the companies for missing the suspicious ones?
I loved this point! It’s so true. They go on about how such and such strategy worked so well for HubSpot. Of course it did, HubSpot has an easy audience: marketers! Turns out those strategies work a lot less well for technical or more discerning audiences.
It takes power away from homeowners when you can’t question the work of a tradesman because you can’t get your own copy of the standards.
It also makes it hard for owner-builders like me who are required to follow the standards but don’t really want to pay for them.
https://www.nationalgeographic.com/science/phenomena/2016/01...
We don’t feel so bad about using paper or wood because it now biodegrades, but if those organisms hadn’t evolved, we’d be worried about wood and paper the same way we worry about plastic.
This article suggests there are some bacteria that has already evolved to eat kinds of plastic:
http://theconversation.com/how-plastic-eating-bacteria-actua...
Maybe in a few decades with more research, we’ll have solutions for plastic that don’t involve sending it across the world?
At first the huge corporations saw OSS as a threat. Their model had been to spend money building software, then charge people for that software. Competing with $0 software scared them and caught them off guard.
Now the corporations and VCs figured out how to use OSS as a weapon. And it’s a deadly one!
The developer insistence on everything being open source means that it’s very easy for a large corporation to destroy a whole lot of ISVs by just releasing an OSS equivalent. VCs know this too.
OSS is used as a PR tool. Projects like React and Angular go a long way to making the parent companies look developer-friendly, even though many developers would normally consider those company’s missions against their values.
OSS is also a powerful way for corporations to commodotise their complements. You see this with the cloud vendors jumping behind Kubernetes to ensure Docker wouldn’t be the next Oracle.
Developers expect everything to be free and open source. We’ll end up in a world where the only programming jobs left will be at the corporations who weaponise OSS the best.
FAANG sponsor open source either for strategic reasons (the Kubernetes example in the blog) or as a goodwill exercise. To the kinds of engineers Facebook would love to hire to help serve ads more devilishly, React goes a long way to making the company look developer friendly. These projects get a ton of marketing support (time to promote the project, brand name, etc) that an all-volunteer nights and weekends grassroots open source project can never complete with.
Then there’s open core, which seems to struggle from the difficulty that open source has a kind of altruistic halo around it, and any attempt to commercialise or restrict parts of it (especially for things that were once open) is seen as bait and switch. Yet to see an Open Core company that isn’t continuiously torn between what it takes to keep the business viable and what it takes to keep the community on their side.
Consulting and support has its own downsides too. It only scales linearly as you can hire people, and the margins are much lower so it takes a lot more people - much less fun of a company to run. And it influences the project roadmap too: “yes the setup process is broken and could do with a wizard, but the complex install process drives a lot of our consulting leads”.
It seems like the huge companies that were threatened by open source have been replaced by huge companies that know how to weaponise open source. Open source still wears a certain halo thanks the nights and weekends code hacker that does it for the love, but open source at any scale is as corrupt - maybe more corrupt - than any proprietary company.
(Disclaimer: my company sells proprietary software but gives customers source access/right to modify and open sources many of the building blocks. As someone who wrote code for the love of it, I often feel like there’s something evil in not making everything open source. But I’m yet to see a less evil open source model)
> Respondent Yana Beklova, who at one stage worked for the Complainant, requested an amount between USD 500,000 and USD 1,000,000 for the transfer of the disputed domain name, that the registrant of the disputed domain name sought to hide its identity through a WhoIs privacy protection service, that the Respondents tried to transfer the disputed domain name out of the reach of the Australian courts to avoid legal proceedings after having received the cease and desist letter, that some of the Respondents named by the Complainant have been involved in previous UDRP proceedings, which suggests a pattern of conduct, and that the disputed domain name was registered to intentionally attract customers looking for the Complainant’s website for commercial gain by creating a likelihood of confusion.
The whole philosophy of StackOverflow was that it wasn’t about sharing opinions, discussions, etc. - it was about questions that have an answer.
Jon’s post points out this difference: that some askers want to ask a question, but haven’t taken the time to understand what makes StackOverflow different to the discussion forums of old.
You can do Linux today actually - you can connect to machines over SSH, run bash scripts, etc. Today there's a Mono dependency, but in the next couple of weeks we'll be rolling out a beta that removes that dependency.
Python scripting support is on the roadmap - :-)
Not that I like the Oracle sales model, but that is impressive. Playing devils advocate, perhaps they are justified to treat sales so well.
The benefit is that you get more consistency between police training across the state and pooled resources. Certainly I think the police are more trusted here.
https://octopus.com/blog/build-your-binaries-once
A real world example of this is that when .NET 4.5 shipped, the compilers in .NET 4.5 for 4.0 code produced different output than the .NET 4.0 compilers would have. So installing a system-level update on Windows on a build server would mean Test and Production got different results:
http://blog.marcgravell.com/2012/09/iterator-blocks-missing-...
Also, releases move through environments at different times:
Monday: 1.0 goes to Test
Tuesday: 1.1 goes to Test
Wednesday: 1.0 goes to Prod
Between 1.0 and 1.1, perhaps you updated Node, or went to Python3 - so your build server had to be updated. Reproducing Monday's build is going to be more difficult the more time that goes by.
As you said, truly reproducible builds are hard. Why not just zip the artifacts and use the same files, instead of rebuilding?
Three areas that I think would have been worth including:
1. Pre-production.
You deploy to test & other pre-prod environments more often than prod. They should use the same scripts/tools/processes as production deployments, only with different permissions.
2. Configuration.
Test and production environments will always have different config settings, so no team will ever be able to deploy to more than one environment without encountering this problem. I think there's still an open question around whether those configuration settings should live in the same source control as the code, in a different source control repository, or a dedicated system. Source control systems and sensitive values (passwords, API keys, etc.) don't always mix.
3. Build your binaries once.
The article is more focussed on dynamic languages, but for compiled languages, I think this is important. If you branch, compile, deploy to test, test it, get the all clear, then compile again and deploy to production, there's a lot of opportunity for differences between what you tested and what goes to production to sneak in.
In fact even for dynamic languages, this might be a valuable practice. What if the JS minifier on one build server is different to another, and the deployed script ends up being different in production to what was tested.
Disclaimer: I'm the founder of Octopus Deploy, and these practices might be biased towards enterprisey .NET/on-premises deployments rather than cloud hosted, dynamic language projects.