It would be funny if there was an exploit in there somewhere, since it seems to be using a completely personal stack and not whatever standard ones they have there
If you were going to run into an exploit it would be on whatever standard stack you run. A personal stack makes creating an exploit much more difficult and requires custom exploit development approach. Is the target worth the effort?
Use a common framework / stack don't update and you will likely be hit by an automated exploit.
I disagree. Using the framework developed by hundreds of talented google engineers will protect you against a large number of attacks. Also, if an attack already exists on the common google framework, then that means google.com is already vulnerable through other usages of the framework, and one more site does not increase your attack surface at all.
There are hundreds of thousands looking for a zero days bugs in bigger frameworks.
None for yours.
Wordpress is a great example. It has better security than your average php blog.
Your average php blog doesn't get attacked by automated exploits attacking known endpoints either.
A determined hacker would find it easier to buy an expliot to a common framework compared to figuring out your custom framework.
We've seen redis and mongo boxes getting hit by no password logins within seconds of going on the public web. Joe's custom cache and database doesn't get hit.
doesn't have my city. bizarre. Brisbane has Bureau of Metrology data back to the dawn of time. Has Sydney, so its not lack of understanding of the AU data
Yea, agreed. Thats why I didn't think it was strange that a given city was missing (though I don't know that Jerusalem or tel Aviv would be crucial to any such list either?)
Pretty neat visualization even though the execution is a bit sluggish
Also weird how it mismatches on the search for "San Francisco" of all places. Looks like it splits the terms on spaces rather than, say, commas, and then matches any of those, so "york rio" will show New York and Rio de Janeiro
Even weirder is that it replaces commas with spaces before splitting the string, so "york,rio" has the same result
I have to assume that the submitted URL won't be around for long, given it looks like an unannounced and unmaintained pet project of somebody who is no longer at the company.
Pretty sure I heard on HN relatively recently that your contract with Google basically explicitly states that any code you write while you work there is considered theirs and they don't distinguish between personal and professional projects. I was surprised to hear this as well (so I'm not asserting that I know this for a fact) but the person who posted this was very explicit that it was true.
So if it is true, then maybe any software engineer that writes some interesting web app trinket who works for Google could have this happen to them. I'm sure the coder who made the above app will show up in these comments in the next couple hours or so lol.
In every employment contract I've ever seen, IP assignment is right there in the dozen or so pages of the contract. It is far too critical to just hide it away in "some company rule book" that the contract references. The contract itself will spell it out in great detail.
Just to double check, if you are developing a project in the evening at your own - Google still owns it(How does it work under CA law)? How about other FAANGs?
If you're in the US, Google claims that they own it (even in California, where local law means they can only do so if it "Relate[s] at the time of conception or reduction to practice of the invention to the employer's business, or actual or demonstrably anticipated research or development of the employer" - they asserted ownership of a Linux driver I wrote on my own time for hardware produced by a company that went out of business before Google even existed).
Google's actually better than many though - there's a process to release your code under an open source license (although with Google's copyright), you can contribute patches to third party open source projects without additional permission (again, under Google's copyright), and you can request that Google assign copyright for a personal project back to you. My understanding is that Apple basically forbid all of this outright. I'm not sure what the situation is at Amazon or Netflix.
two years ago at an apple all hands someone asked craig federighi about this and he more or less said "i know you all hate this {we get a lot of questions about this issue}... but for legal reasons we're not going to change it."
What does it mean to contribute to third party OSS (that has whatever X licence) under Google's copyright?
That the project owners couldn't change licence (and keep that code) without Google's consent? Is this why some (/all of a certain size or commerce) worry about CLAs?
Google remains the copyright holder of the submitted patch. There's nothing special in this from the project's perspective - in the absence of copyright assignment as a required part of patch submission, the default is that the submitter retains copyright. In this case, it's just Google claiming ownership of their employee's patch rather than the employee holding the copyright.
Microsoft has the most permissive moonlighting policy amongst the major SE companies that I’m aware of. Sometimes they actively encourage it, like during the goldrush days of the Windows Phone and Windows 8 App Stores when they had on-campus hackathons for personal projects and everyone there was using their company-issue laptops, but I guess they were just desperate for content, heh. I remember we had received blanket assurances in writing that MS wouldn’t try to co-opt anything we’d post, just so that we don’t do anything that competes with Office…)
Thing is while many cases are easy to tell if they’re obviously in violation of policy, it’s still possible to sleepwalk into it:
Doing personal email in OWA/Gmail on a company laptop in lunch? OK
Doing personal email in desktop Outlook on a company laptop? OK (as was my understanding of policy at the time: there was a carve-out for having personal email on company hardware)
Doing personal email which involves a quick code-review of a rando’s PR to your personal github repo? hmmmm…
What about making slight modifications to that code?
What if you clone the repo because you need to get a bigger-picture?
What if you inadvertently start a build of your cloned repo?
I doubt Apple would make that claim of your personal Macbook. Likewise if you're not using the /Microsoft account on GitHub, it's not 'company equipment' in that sense, it's just happens that in a personal capacity you are a client of (a subsidiary of) the company you work for.
In general to claim personal projects as your own, in California anyway, I’ve been told it has to be on your own time and on your own equipment. (and probably a non-conflicting subject matter) By offering hosting for the personal time projects on their infrastructure, Google would make it their equipment.
A caveat is that, in California, you own your project provided that (1) you didn't use any company resource, (2) you didn't do it on work hours, and (3) your project is not related to anything that the company does. Of course, in case of Google, it does pretty much everything, which means Google can easily claim anything you do as its own. (IANAL, don't quote me on this.)
obviously you should read the contract before signing with $MEGACORP_THAT_PAYS_YOUR_BILLS, and I agree with (1) and (2) and (3) kinda sorta not really since it basically just implies you are a work machine, but it just seems so rife with the potential for abuse.
Thankfully I haven't heard of many other cases of this getting people in too much trouble other than Serge Aleynikov and his case with GS (and honestly, it was stupid of him to upload even open source code from work computer to a remote server).
I worked at GS and it was the most locked down development environment ever, so it wouldn't even be feasible/productive for you to "accidentally" work on side projects on these machines. Most security/audit trainings were around "Really - if you do this, you will instantly be teleported into the center of the nearest neutron star"
IIRC, he saw some open source code used in Goldman Sachs, believed that they should be open (whether right or wrong, I don't know), and uploaded it public. Right? I remember HN discussion from years ago.
I don't think his story is related to developers' side projects. If you work for Company X and knowingly upload some of its internal repository to a public place, of course you're going to be fired. Would you expect anything else?
from my understanding it was more nuanced than that - the code was already open source but he then made some changes to the source and wanted to keep those changes. Obviously
this is pretty bad too, but eventually all charges were dropped and this might have just been because how aggressive the plaintiffs and the FBI goons were in pursuing this guy, especially after considering the accusations of "stealing code" that is so dangerous it could manipulate the fabric of financial markets.
In any case, I just pulled up my GS NDA and I probably shouldn't comment any further.
Funny side note though: GS realtime risk/lazily loaded object graph database is called SecDB, and @saleyn has a repo on GitHub called secdb which has to do with having an efficient data store for financial data... I guess he was feeling invulnerable after getting off of the previous charges haha
I live in the US, but not in California. I've worked at three different companies. Each time when signing the employment contract I have pushed back on any terms that would make code that I write outside of work owned by the company. I haven't had any push back on this. My experience may be limited and may not generalize, but I find that in most employment contracts there are just a handful of points that the other party will not concede on. This may be less true at $MEGACORP.
Neither of the two $MEGACORPs (or at least $LARGECORPs) I've worked at have had any issue with various things I've done on the side and I've never kept them secret. So it varies--and neither were in California.
When I took my current position at a financial firm, it had the clause, but it also came with a form where I could declare, in specific or broad terms, things I did outside of employment. The company still included language affirming the CA guidance on this, and I spoke to HR and gave them hypotheticals, which they responded to in email.
I have lots of technical interests that aren't even remotely related to fintech, so I feel comfortable with these boundaries. Plus, at work I'm given a Windows machine, and all my personal projects -- and other commercial endeavors -- won't run on Windows. For that reason alone, I haven't requested alternative hardware/OS even though I could.
Anything you do on your own time, with your own equipment, that is not related to your employer’s line of work is yours, even if the contract you signed says otherwise
(from California state law, but it could reasonably be directly in a contract from a reasonable employer)
Joel makes the case that if you work for Google, this is meaningless because what isn't related to Google? They do everything. This makes sense to me, and in Google's defense, they pay the part. But then Joel goes on to make the case that "related" is such a vague term that as an employee you can't safely depend on it, even if you work for a company with a narrower scope (ie nearly any other company).
I don't know, this seems pretty far fetched to me. Say, you work at SoundCloud and your side project is an iOS game. Wrote it on your own device, in your own time, and so on and so forth. That is so obviously not related that I have a very hard time seeing how, with a clause such as this, any potential judge or jury could decide otherwise. Let alone the reputation damage SoundCloud could get trying to lawyer up on this shit. It just makes no sense.
Therefore, I still feel it's a draconian piece of contract and that, unless you apply at a company that truly does everything or pays top dollar, it shouldn't be a reasonable expectation to sign stuff like that.
How does this work for contributions to FOSS? What happens if $MEGACORP decides that any off-work contributions to FOSS software are rightfully theirs?
It would work in the same way as any off-work side projects. If it's (legally) your code, then you can do whatever you want with it, whether you sell it or contribute to FOSS. Otherwise, it was never your code to begin with, so you couldn't "contribute to FOSS" any more than you could donate your work notebook to a charity.
In reality, I think Google has an "open source committee" where you would show your proposed contribution, get approved, and then you can contribute to outside open source projects. I heard the process was fairly reasonable.
Just to add to the other comments here, from a more freelance perspective:
Most work for large corporations is done under a "work for hire" agreement which means the company owns the copyright to the source you write. Additionally, many big companies will pressure you to accept employment clauses that claim ownership of anything you create in your free time, 24/7. This is not just for code, it's also standard for contract employees in the TV and film industry.
Note I'm distinguishing here between "employee" and "freelancer". I refuse to sign work-for-hire agreements, which means when hired, I still own the copyright to the source. I grant the company hiring me free use and modification and in some cases, right to resell.
I had a case of a startup where I was the sole coder, who turned around and sold their app to [Fortune 100 company] who didn't realize that I still owned this source. This company then asked me to consult - offering half what I'd been paid to write it - and further, asked me to sign a contract stating that they owned the code and everything else I made while I was consulting with them. Oh and also, without my help, they were missing some major assets they needed to recompile it.
I told them where to stick it, so they hired a bunch of people to try to reverse engineer it rather than agreeing to my terms. That didn't work (not least because I had trapped the fuck out of the code they did have, in ways that made it crash the compiler) so they ended up down a couple million bucks and shuttering the site. If they had only been nice, and hadn't come at me with the aggressive attitude, it could've been a nice little asset.
If you do not sign a work-for-hire agreement, and you're an independent contractor - not an employee - you own the source code unless otherwise stated in a contract. Beware that some jurisdictions will view you as a de facto employee in some cases (for better or worse). And my advice to anyone listening is, never, ever sign a contract with one of these companies that stipulates that they own your off-time work product. It's completely abusive and unnecessary. It's not worth taking a job with such a company.
If by the place I work at you mean nowhere, yes I own a lot of equity there. But in my past employment and contracting relationships, equity or not, I can't imagine shipping code with kill switches or booby traps that weren't in the spec or otherwise requested by the person I was working for.
To be fair, the kill switches were mostly to prevent decompilation and reverse engineering. And I added them on my own time after I learned of the pending sale, which no one had consulted me about, and before the company took control of the servers, which I was administering. Was it worth my time? Absolutely.
The people who sold the company intentionally misrepresented what they actually owned. They only owned the final compiled software they had purchased, and the right to use and repair or modify it. They didn't have a right to sell the source code, and I had no obligation to make it easy for the buyer to reverse engineer it.
Is that even a lesson that needs to be learnt? Nuts not to, surely? Even this sort of issue aside, what happens when you want to add a feature and for whatever reason (that could include the original author's retirement or death!) hire someone else?
Binary delivery seems like buying some cool-looking alien device that manufacturers widgets; you have no idea how it works or how to maintain it, so you just sell widgets hoping it keeps working without maintenance.
Imagine you're a company looking for software as a cloud service. Suppose the best software for the purpose only does 70% of what you want. So you contact the company who runs that cloud service and pay them to develop the additional features and specific business logic you need. Then you contract with them to manage private servers with your build on them. At no point have you seen or purchased source code - not even for the 30% new features that were written on your behalf. Those are entwined with, and only work as a result of, the other 70%. You're basically using a black box that's maintained by another company, and trusting in that company's competence and goodwill. That's essentially what the situation was here. If you sold your company and touted your special build of the software as a feature, and passed the contract along to the buyer, that would be fine. But if you misrepresented the situation and said you owned 100% of the code and were selling that along with the right to distribute it, that would be fraud.
Of course, you could have gone the more expensive route and hired your own team of coders to write the whole thing from scratch, under whatever license you wanted, and then you would own the source. But if a company needs document software or meeting software in the cloud, how many pay to reinvent the wheel and roll their own?
Perfect! Possession is nine-tenths of the law. Pacta sunt servanda. You made a good deal about something you created, and you had the powerful means to enforce the deal. I like that. People can say this "sounds unethical", but this is exactly what every other employee or freelancer should aspire.
Yes, it's nice to have it _when you need it_, but consider all your other honest customers that years later discover your kill switches littered throughout the code they paid you to write. They'll have to pay some other developer to untangle or work around them. If your contract with your customer doesn't include a provision for the them, or it wasn't on the roadmap you worked out with your customer, you're delivering them something with defects that may cost them later. To say nothing about how they will feel about you once they discover you anticipated that they were a cheat the whole time you worked with them.
Paranoia-driven engineering decisions like this are no good. If your customers are bottom-of-the-garbage-can customers, get better customers. If you can't find better customers, then perhaps you are right where you deserve to be, doing garbage work for lousy cheats, which you must booby trap to make sure you get paid. If that's where you are, fine, but I hope no one with a brighter future reads these kinds of comments and thinks that being a bottom feeder is the norm.
I might have misread the comment. I was under the assumption that a business deal went wrong, and that the programmer never ceded ownership of the source code to begin with.
Ofcourse, preventing is better than healing. But if you happen to make a deal and the other party turns out to be illegally selling something that's yours, than it is better to be able to enforce that deal.
I think you may be misunderstanding my position. To be clear, the original company owned the functionality they were purchasing. They needed software that did X, Y and Z. A lot of the code that went into it was code that I had already written for other purposes, including my own projects. They never owned that. They then sold the company as if they owned it - i.e. as if the purchaser would have free access to the source to repackage as they liked. I had no such contract with the original company, much less the buyer.
I should also clarify that this was a PHP backend that was handed over freely, and a public-facing Flash app that was handed over as an SWF. At heart it was a highly specialized video / social platform. Graphics assets were handed over. What was not handed over were the front-end AS3 scripts; and I used a few hex editing methods and other ways to trap the compiled bytecode that would crash all known SWF decompilers. I also had it phone home if someone tried to decompile it. (All those decompilers would start by importing the SWF onto the root object of another SWF; root detects a higher parent; kaboom).
What those guys did, and what the other company tried to do to me, was unethical. I didn't even have the right to sell them exclusive use of some of that source, because it was already deployed in other commercial projects. I didn't have the right to sell it out from under the other companies who already had paid for use of it. My refusal to sign [Fortune 500 company's] contract was on their behalf as well as my own. That company wouldn't listen to reason, so I left them a working binary and said "good luck", and then watched it keep phoning home until they gave up trying to decompile it.
Also, I've never prevented an honest customer from accessing source code they needed, and in this case I was willing to negotiate a non-exclusive use of the source code with the new company, for a fair price. But they refused that. They demanded sole ownership for $0, and they wanted to pay me half my rate for consulting. Not to mention they themselves are one of the most blatantly unethical companies in the world. I didn't feel like I owed them anything.
Here in the UK I worked for a small company in 2001. Each person recorded time spent on projects on one spreadsheet on a shared drive, a nightmare. I had a free website with uklinux (now no more) and in my spare time quickly knocked together a web based solution. Over a year or two I enhanced it, in my own time, and everyone liked it. But the company started losing money and one day the finance director came up to me and said my contract said that the system belonged to the company and I should hand it over. Well it was on my web site - I said no. Next day people complained they couldn't log in. I sent them to the finance director! Oh, and I had explained to the systems guys how to do a daily download of their data, so that was OK wasn't it? Next day finance director changed his mind.
The company did go bust, and only then did I open it up as a paid service. I ran it as a side project, but got a steady profit from it. Closed it when I retired.
> What position do you have to be in at Google to be able to upload your side projects to "google.com/X"?
Software Engineer.
You just need to pass code review to make such a change - convincing people reviewing it would be the tricky part but not impossible. You could brand it a "20% project" or something.
There's actual plenty of interesting (usually old) content that's obviously not an official project hosted at this kind of URLs. It's a bummer you can't use Google the search engine to discover them.
Pretty sure Google Gulp is an old april fools joke they kept up. Google Moon makes sense, unfortunately it looks like it hasn't been kept up to date and the resolution if you zoom in isn't the best.
This would be a good example of the true definition of the "dark web", right? Websites that exist but are not accessible from any standard search engine or hosted directory.
it's a rare view to see a site that contains 97.4% html. I can only imagine what's like to build a newer version right now and if it's going to enhance UX or just add a ton of code and unnecessary logic.
Maybe it was a pet project of someone from early google and it's some kind of easter egg type thing. It could have been moved to github 4 years ago, but it was written a long time ago. Seems like he is still at the company:
https://github.com/lmanul
Came here to mention weatherspark. I had thought they completely died out when they abandoned their awesome flash UI - good to see they still have nice data and visualizations.
Weatherspark doesn't seem to have a summary page of the monthly climate around the world. Seems like on Weatherspark you need to visit dozens of pages to get an idea of the best place to travel to.
>When you click on a month it basically sorts all cities by order of "comfiness" which is a simple combination of temperature and precipitation (not too hot, not too cold, not too rainy).
Yeah, by the name of it I thought it was going to be some alt version of TSP where you have to go around the world and reach every city on the list while always having sunny days.
This would be even more useful if I could search my ideal temp range, and then it would highlight only the periods where a given city meets the requirement.
(Currently in lockdown here in Sydney so all I can do is dreaming of travelling... )
Why is this #1 on HN? Just because of the high-profile URL? The project itself seems really basic, or am I missing something? Also there's a bug when search query contains a space...
I don't mean to come off as super negative. I hate how the Internet is so overly critical. And Twitter is just the worst. I'm seriously asking if I'm missing something. THanks!
Just in case it's of interest to you, it's very easy to get historical weather data from weather stations all over the US [0] (and I assume most EU countries also publish this data).
I was about to say - I took this for a Gödel, Escher, Bach reference given it's someone at Google, as well as the fact the pun is very much Hofstadter's style. In fact I would go so far as to say that the whole project was made to justify the title.
To add to the conspiracy, what if Google really does have a bin of pet products laying around and whips them out on social media at just the right time? Right now, as of me posting this comment, this[0] is one of the top posts on HN. Google's PR team probably knows how to game high traffic sites at the right times with artificially inflated scores.
I'd like to see something like this combined with daylight hours. When daylight savings ends in November and twilight starts being as early as 5pm the lack of sunlight makes me feel pretty gloomy. I find myself much happier when it's still a bit bright outside at 7pm. Good weather is nice but I think daylight affects my wellbeing a lot more!
I dream of someday being able to spend those couple months of "it's pitch black outside at 6pm" somewhere else in the world with more sun.
Is this in California? Never occurred to me that this is the case there:)
I’ve spent a fair bit of time in Middle East and it is sunset at 6-7pm pretty much throughout the year. And I must say when I traveled north to Netherlands/Germany in the summer - the experience was almost psychedelic.
My dream is to be able one day to travel between north and south to follow longer day time(and summer) throughout the year.
> I’ve spent a fair bit of time in Middle East and it is sunset at 6-7pm pretty much throughout the year
That's mostly about where you are within your timezone and whether you have Daylight Savings Time. On local solar time the sun has to spend as many days per year setting before 6pm as after.
It would be amazing if they could choose whether to show °F or °C based on the country I'm accessing the website from instead of choosing one as a default.
The code does check your browser locale of "en-US" and sets it to the "wacky" unit (Fahrenheit). Given that it's plain vanilla JS and only some rudimentary XMLHttpRequest, I'm not sure it can do much more?
Seems to skew hot. Sydney is shown as the cold months blue and hot months still green, yet its a city where the summer heat is far worse than winter cool.
Thanks for the ability to have a comma-separated list of locations. Such a trivial and useful thing. None of the alternatives posted in this thread seem to have this.
182 comments
[ 2.9 ms ] story [ 189 ms ] threadUse a common framework / stack don't update and you will likely be hit by an automated exploit.
None for yours.
Wordpress is a great example. It has better security than your average php blog.
Your average php blog doesn't get attacked by automated exploits attacking known endpoints either.
A determined hacker would find it easier to buy an expliot to a common framework compared to figuring out your custom framework.
We've seen redis and mongo boxes getting hit by no password logins within seconds of going on the public web. Joe's custom cache and database doesn't get hit.
a) 10,000 users of a common framework + 1 user of a custom framework.
and
b) 10,001 users of a common framework
I contend that (a) has a larger attack surface than (b).
Also weird how it mismatches on the search for "San Francisco" of all places. Looks like it splits the terms on spaces rather than, say, commas, and then matches any of those, so "york rio" will show New York and Rio de Janeiro
Even weirder is that it replaces commas with spaces before splitting the string, so "york,rio" has the same result
It's cool nontheless, would love to hear how that came together, not something you expect to just be in a google.com subfolder.
I have to assume that the submitted URL won't be around for long, given it looks like an unannounced and unmaintained pet project of somebody who is no longer at the company.
This seems to have been a year ago at most judging commits, so can't be something that's been there for ages.
So if it is true, then maybe any software engineer that writes some interesting web app trinket who works for Google could have this happen to them. I'm sure the coder who made the above app will show up in these comments in the next couple hours or so lol.
EDIT: Omg it's the cartoonist guy!
In every employment contract I've ever seen, IP assignment is right there in the dozen or so pages of the contract. It is far too critical to just hide it away in "some company rule book" that the contract references. The contract itself will spell it out in great detail.
Google's actually better than many though - there's a process to release your code under an open source license (although with Google's copyright), you can contribute patches to third party open source projects without additional permission (again, under Google's copyright), and you can request that Google assign copyright for a personal project back to you. My understanding is that Apple basically forbid all of this outright. I'm not sure what the situation is at Amazon or Netflix.
https://opensource.google/docs/iarc/
That the project owners couldn't change licence (and keep that code) without Google's consent? Is this why some (/all of a certain size or commerce) worry about CLAs?
That doesn't mean that they won't try to sue you though.
Thing is while many cases are easy to tell if they’re obviously in violation of policy, it’s still possible to sleepwalk into it:
Doing personal email in OWA/Gmail on a company laptop in lunch? OK
Doing personal email in desktop Outlook on a company laptop? OK (as was my understanding of policy at the time: there was a carve-out for having personal email on company hardware)
Doing personal email which involves a quick code-review of a rando’s PR to your personal github repo? hmmmm…
What about making slight modifications to that code?
What if you clone the repo because you need to get a bigger-picture?
What if you inadvertently start a build of your cloned repo?
Where is the line?
A caveat is that, in California, you own your project provided that (1) you didn't use any company resource, (2) you didn't do it on work hours, and (3) your project is not related to anything that the company does. Of course, in case of Google, it does pretty much everything, which means Google can easily claim anything you do as its own. (IANAL, don't quote me on this.)
Thankfully I haven't heard of many other cases of this getting people in too much trouble other than Serge Aleynikov and his case with GS (and honestly, it was stupid of him to upload even open source code from work computer to a remote server).
I worked at GS and it was the most locked down development environment ever, so it wouldn't even be feasible/productive for you to "accidentally" work on side projects on these machines. Most security/audit trainings were around "Really - if you do this, you will instantly be teleported into the center of the nearest neutron star"
IIRC, he saw some open source code used in Goldman Sachs, believed that they should be open (whether right or wrong, I don't know), and uploaded it public. Right? I remember HN discussion from years ago.
I don't think his story is related to developers' side projects. If you work for Company X and knowingly upload some of its internal repository to a public place, of course you're going to be fired. Would you expect anything else?
In any case, I just pulled up my GS NDA and I probably shouldn't comment any further.
Funny side note though: GS realtime risk/lazily loaded object graph database is called SecDB, and @saleyn has a repo on GitHub called secdb which has to do with having an efficient data store for financial data... I guess he was feeling invulnerable after getting off of the previous charges haha
It's overly broad so that if you have a dispute about something you did outside of work they don't have to prove shit in court to get you to stop.
I have lots of technical interests that aren't even remotely related to fintech, so I feel comfortable with these boundaries. Plus, at work I'm given a Windows machine, and all my personal projects -- and other commercial endeavors -- won't run on Windows. For that reason alone, I haven't requested alternative hardware/OS even though I could.
It’s just so in conflict with the creative process.
https://www.youtube.com/watch?v=eiyfwZVAzGw
Anything you do on your own time, with your own equipment, that is not related to your employer’s line of work is yours, even if the contract you signed says otherwise
(from California state law, but it could reasonably be directly in a contract from a reasonable employer)
Joel makes the case that if you work for Google, this is meaningless because what isn't related to Google? They do everything. This makes sense to me, and in Google's defense, they pay the part. But then Joel goes on to make the case that "related" is such a vague term that as an employee you can't safely depend on it, even if you work for a company with a narrower scope (ie nearly any other company).
I don't know, this seems pretty far fetched to me. Say, you work at SoundCloud and your side project is an iOS game. Wrote it on your own device, in your own time, and so on and so forth. That is so obviously not related that I have a very hard time seeing how, with a clause such as this, any potential judge or jury could decide otherwise. Let alone the reputation damage SoundCloud could get trying to lawyer up on this shit. It just makes no sense.
Therefore, I still feel it's a draconian piece of contract and that, unless you apply at a company that truly does everything or pays top dollar, it shouldn't be a reasonable expectation to sign stuff like that.
Especially in the rare case that the game becomes popular.
In reality, I think Google has an "open source committee" where you would show your proposed contribution, get approved, and then you can contribute to outside open source projects. I heard the process was fairly reasonable.
Most work for large corporations is done under a "work for hire" agreement which means the company owns the copyright to the source you write. Additionally, many big companies will pressure you to accept employment clauses that claim ownership of anything you create in your free time, 24/7. This is not just for code, it's also standard for contract employees in the TV and film industry.
Note I'm distinguishing here between "employee" and "freelancer". I refuse to sign work-for-hire agreements, which means when hired, I still own the copyright to the source. I grant the company hiring me free use and modification and in some cases, right to resell.
I had a case of a startup where I was the sole coder, who turned around and sold their app to [Fortune 100 company] who didn't realize that I still owned this source. This company then asked me to consult - offering half what I'd been paid to write it - and further, asked me to sign a contract stating that they owned the code and everything else I made while I was consulting with them. Oh and also, without my help, they were missing some major assets they needed to recompile it.
I told them where to stick it, so they hired a bunch of people to try to reverse engineer it rather than agreeing to my terms. That didn't work (not least because I had trapped the fuck out of the code they did have, in ways that made it crash the compiler) so they ended up down a couple million bucks and shuttering the site. If they had only been nice, and hadn't come at me with the aggressive attitude, it could've been a nice little asset.
If you do not sign a work-for-hire agreement, and you're an independent contractor - not an employee - you own the source code unless otherwise stated in a contract. Beware that some jurisdictions will view you as a de facto employee in some cases (for better or worse). And my advice to anyone listening is, never, ever sign a contract with one of these companies that stipulates that they own your off-time work product. It's completely abusive and unnecessary. It's not worth taking a job with such a company.
Bellum omnium contra omnes
Usually the developer gets taken advantage of because they assumed the world to be fair and equal and companies to have your best interest.
In this case the developer got wind of an attempt to take advantage of him. He put in protection to prevent this. That's what a company would do.
To not understanding why someone needed do that strikes me as a bit naive.
The people who sold the company intentionally misrepresented what they actually owned. They only owned the final compiled software they had purchased, and the right to use and repair or modify it. They didn't have a right to sell the source code, and I had no obligation to make it easy for the buyer to reverse engineer it.
Binary delivery seems like buying some cool-looking alien device that manufacturers widgets; you have no idea how it works or how to maintain it, so you just sell widgets hoping it keeps working without maintenance.
Of course, you could have gone the more expensive route and hired your own team of coders to write the whole thing from scratch, under whatever license you wanted, and then you would own the source. But if a company needs document software or meeting software in the cloud, how many pay to reinvent the wheel and roll their own?
Paranoia-driven engineering decisions like this are no good. If your customers are bottom-of-the-garbage-can customers, get better customers. If you can't find better customers, then perhaps you are right where you deserve to be, doing garbage work for lousy cheats, which you must booby trap to make sure you get paid. If that's where you are, fine, but I hope no one with a brighter future reads these kinds of comments and thinks that being a bottom feeder is the norm.
I should also clarify that this was a PHP backend that was handed over freely, and a public-facing Flash app that was handed over as an SWF. At heart it was a highly specialized video / social platform. Graphics assets were handed over. What was not handed over were the front-end AS3 scripts; and I used a few hex editing methods and other ways to trap the compiled bytecode that would crash all known SWF decompilers. I also had it phone home if someone tried to decompile it. (All those decompilers would start by importing the SWF onto the root object of another SWF; root detects a higher parent; kaboom).
What those guys did, and what the other company tried to do to me, was unethical. I didn't even have the right to sell them exclusive use of some of that source, because it was already deployed in other commercial projects. I didn't have the right to sell it out from under the other companies who already had paid for use of it. My refusal to sign [Fortune 500 company's] contract was on their behalf as well as my own. That company wouldn't listen to reason, so I left them a working binary and said "good luck", and then watched it keep phoning home until they gave up trying to decompile it.
Also, I've never prevented an honest customer from accessing source code they needed, and in this case I was willing to negotiate a non-exclusive use of the source code with the new company, for a fair price. But they refused that. They demanded sole ownership for $0, and they wanted to pay me half my rate for consulting. Not to mention they themselves are one of the most blatantly unethical companies in the world. I didn't feel like I owed them anything.
The company did go bust, and only then did I open it up as a paid service. I ran it as a side project, but got a steady profit from it. Closed it when I retired.
Software Engineer.
You just need to pass code review to make such a change - convincing people reviewing it would be the tricky part but not impossible. You could brand it a "20% project" or something.
http://google.com/googledance2004 http://google.com/moon http://google.com/googlegulp http://google.com/heart
"No results."
I'm told that that's because who knows how that site works is still with the company.
So Google blocks it’s own content from being indexed?
He certainly had a free reign at Google.
I was expecting the top temperature legend to let you filter locations by temperature, but that's not the case.
"Org charts" comic by Manu Cornet
I never realized it the org charts was created by the same dude.
Has all the breadth, and much more depth.
Arequipa is even better.
I think most comfortable places lie on west shores at medium latitude.
Weatherspark doesn't seem to have a summary page of the monthly climate around the world. Seems like on Weatherspark you need to visit dozens of pages to get an idea of the best place to travel to.
>When you click on a month it basically sorts all cities by order of "comfiness" which is a simple combination of temperature and precipitation (not too hot, not too cold, not too rainy).
[0] https://twitter.com/lmanul/status/924470796264751104
https://github.com/lmanul/welltemperedtraveler
https://en.m.wikipedia.org/wiki/Negative_temperature
(Currently in lockdown here in Sydney so all I can do is dreaming of travelling... )
https://championtraveler.com/travel-weather-map/
Allows you to choose and filter weather preferences then tells you where you can likely travel to find that weather.
Now if only he could get it published on Google.com/...
I don't mean to come off as super negative. I hate how the Internet is so overly critical. And Twitter is just the worst. I'm seriously asking if I'm missing something. THanks!
[0] https://www.ncei.noaa.gov/products/land-based-stations
https://en.m.wikipedia.org/wiki/The_Well-Tempered_Clavier
Kinda neat to see how much attention such a URL can draw to an otherwise-unremarkable page.
[0] https://news.ycombinator.com/item?id=27841963
I dream of someday being able to spend those couple months of "it's pitch black outside at 6pm" somewhere else in the world with more sun.
I’ve spent a fair bit of time in Middle East and it is sunset at 6-7pm pretty much throughout the year. And I must say when I traveled north to Netherlands/Germany in the summer - the experience was almost psychedelic.
My dream is to be able one day to travel between north and south to follow longer day time(and summer) throughout the year.
That's mostly about where you are within your timezone and whether you have Daylight Savings Time. On local solar time the sun has to spend as many days per year setting before 6pm as after.
You are more likely to be in a country that has banned anti-personnel land mines than one using Fahrenheit.
They should be deriving colours from a thermal comfort range plotted on a psychrometric chart for example https://i2.wp.com/theenvironmentalarchineer.com/wp-content/u...
http://infocampus.co.in/ui-development-training-in-bangalore... http://infocampus.co.in/web-designing-training-in-bangalore.... http://infocampus.co.in/web-development-training-in-bangalor... http://infocampus.co.in/react-js-training-in-bangalore https://www.firstenquiry.com/reactjs-training-in-bangalore http://infocampus.co.in/angularjs-training-in-bangalore.html http://infocampus.co.in/angulartwo-training-in-bangalore.htm...