Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The Cathedral and the Bizarre

marktarver.com

Post reply on HN
Print view View original

101–110 of 130 posts

Re: The Cathedral and the Bizarre

#101

Post by vinay_ys »

codyb wrote: post #77
perl4ever wrote: post #63

Earlier quoted context omitted.

Generally, expectations do not define reality. I'm not just being contrary, I've worked on a project for a year or two that got scrapped, and I think it happens a lot. Another way to look at it, is that if you have programmers that are regular employees and not temps or something, you might as well have them work on something that's not very important, or that's risky, when you don't have something better for them to…

I’m convinced if they stopped paying us all to rewrite things that work the market would fall out from under us. Every job I’ve ever been involved with has always had some code base that was in framework X and now needs to be in framework Y, or on in house hardware and needs to move to the cloud, etc etc. And having just interviewed every place has a project like that. “We were using PHP for the mvp but are moving to…

Most software engineers I know love to build software for the sake of itself. They are not all that interested in the business. Most businesses for which software is being produced is not personally relatable or interesting to most software engineers.

To keep engineers motivated, the work has to have sufficient amount technical accomplishments and not just churning out business features.

Luckily, due to continuous growth in market and scale of users, and improvements in hardware, there is opportunity for serious amount of technical work in any project.

Also, faster time to market is of real business value. Compromises are done in ideal perfect designs to meet timelines (or resource constraints) – hence, tech debt occurs and it is real and just as useful as financial debt to help scale a business in a competing marketplace.

And speculative/iterative market feedback based development is also real. All these things add up to one thing – things that were right for yesterday isn't right for tomorrow.

So we have to rewrite or refactor software continuously. And that's perfectly normal. It is actually one of the biggest strengths of software based solutions – that we can change things live - while it is already in use - change the engines of the rocket while it is flying - so to speak. We have to embrace this and leverage it disruptively.

Those who pine for perfect software are denying realities of the larger real-world.

Re: The Cathedral and the Bizarre

#102

Post by livueta »

sdenton4 wrote: post #74
joppy wrote: post #36

Earlier quoted context omitted.

Bourbaki were certainly made up of a collection of leading mathematicians, but the works of Bourbaki are not innovations themselves. As much as I love the books, they do not contain major new results, rather they are an excellent organisation of pre-existing knowledge, perhaps with some novel proofs.

Meh. Bourbaki is merely indicative... Someone who believes that only individuals can produce theorems hasn't really worked in mathematics. It's a no true Scotsman argument.

I guess Terry Tao would be a good example of collaborative mathematics? He was mentioned in another thread and I noticed the following while reading his wiki page:

> Within the field of mathematics, Tao is known for his collaboration with Ben J. Green of Oxford University; together they proved the Green–Tao theorem. Known for his collaborative mindset, by 2006, Tao had worked with over 30 others in his discoveries,[15] reaching 68 co-authors by October 2015.

Re: The Cathedral and the Bizarre

#103

Post by vinay_ys »

Writing software within old-school enterprises is a daunting task involving lots of proprietary systems that create a lock-in of sorts. Complex build and test systems, complex deployment and production systems are the norm. Weird tools and libraries with their own proprietary names and conventions increase the learning curve for any new-joinee. Some even have very proprietary non-standard language, runtime etc. making the skills acquired in such environments to be largely useless outside of such environments.

Open source culture changed all this drastically. This culture started with individuals loosely connected via Internet who collaborated on writing software and without much complex central infrastructure. Common conventions and standards emerged as these individuals openly discussed them and took with them to other projects – not just conventions and standards, but also the code. Without the enterprise lock-in, this has proven to be extremely successful and cost-effective.

Most commercial software companies today follow this new model for their internal software development practices. Using openly available technology components lets everyone to acquire portable skills that they can continue to improve throughout their careers across many different companies. This drastically reduces the cost of software production as cost of acquiring skills is amortized over many companies and cost of development effort of these open source standardized components is also amortized over many corporations/projects.

Every company I worked in has supported contributing to open-source with the selfish motive to attract talented engineers. And the engineers like contributing to open-source projects as a way of showing off their skills and building their resume.

Re: The Cathedral and the Bizarre

#104

Post by itronitron »

peterwwillis wrote: post #88

Old, large tapestries were often gigantic, complex, expensive, time-consuming, exquisitely intricate carpets that hung on walls. They could take from years to decades to complete, and consume fortunes. One of their primary functions was to show off how wealthy or important you were. Now that we have machines that can do weaving and sew patterns automatically, it's possible to create a tapestry without as much time an…

well, there are tools for auto-generating code from schema files but no machines yet for creating whole applications. Based on my experience, I think the last frontier will be mapping the user-possibility-workflow space to action-consequence-application-state space.

Re: The Cathedral and the Bizarre

#105

Post by BlueTemplar »

gliese1337 wrote: post #73
ThrowawayR2 wrote: post #58

Earlier quoted context omitted.

Generally, a business pays developers to write software that it expects to be put to immediate use, which means that most of the time it functions, at least minimally. I'm sure exceptions exist, of course. FLOSS abandonware has no such constraint.

My first six months of employment at my current job consisted of going from one project to another that the C-levels listed as "a top priority for the company"--like, called all-hands meetings and told everyone about this great thing and how important the people working on it were--and then decided to cancel. Two years later, that still happens from time to time. So, from my point of view, it looks much more like bus…

Honestly, isn't it bothering you that they are wasting your time ?

Re: The Cathedral and the Bizarre

#106

Post by zimpenfish »

buzzkillington wrote: post #44
doublement wrote: post #9

Earlier quoted context omitted.

Moreover he's using the large amount of wreckage to argue that the Darwinian process of selecting "category killers" isn't working, which is completely backwards. If "category killer" is a winner-take-all competition, then there are going to be a vast number of losers, and we do see a vast number of unmaintained projects. It's just that no one is bothering to remove them from view when they're defeated.

I'm unaware that the goal of evolution is to have one lion win the lion competition and eat all other lions. Maybe we should stop using bad metaphors? Writing software is not a Darwinian process, neither is competition on the market place. Unless you got your OS by randomly twiddling bits in the binary file, in which case why are we even talking about open source?

> have one lion win the lion competition and eat all other lions

If that improved the chances of that lion's genes surviving, it would absolutely be selected for. Obviously it generally doesn't because there's safety in numbers, etc.

> the goal of evolution

Minor nit: evolution has no goal. It is undirected, random, capricious, etc.

Re: The Cathedral and the Bizarre

#107

Post by BlueTemplar »

Terretta wrote: post #67
doublement wrote: post #9

Earlier quoted context omitted.

Moreover he's using the large amount of wreckage to argue that the Darwinian process of selecting "category killers" isn't working, which is completely backwards. If "category killer" is a winner-take-all competition, then there are going to be a vast number of losers, and we do see a vast number of unmaintained projects. It's just that no one is bothering to remove them from view when they're defeated.

The analogy might be clearer if instead of GitHub we used Fossil.

He does talk about how huge corporations use open source and exploit open source developers for their own profit...

Re: The Cathedral and the Bizarre

#108

Post by benibela »

sildur wrote: post #18
sdenton4 wrote: post #17

"No mathematical theorem, no enjoyable novel, no work of art of any importance, have ever been produced by a herd. I fail to see why innovative software ought to play by a different set of rules." Nobody tell this guy about Bourbaki... He's also probably never enjoyed a movie.

A novel also is produced by a “herd”. There are a lot of people involved. People that make suggestions, corrections, the editor... It’s not a single person effort.

However, you can still read a novel the same after someone else makes corrections or add things, even if the new things are not good.

Software can become complete broken by a bad patch.

That is seen the best at security issues. Like someone thinks it is a good idea to add a heartbeat extension to ssl. A basically pointless extension, if you have a novel with a pointless appendix from an editor, basically no one would read the appendix and it does not matter that it is there. But with software it ruins the entire server security completely

It is better to pass on 100 good patches than risking one bad patch

Re: The Cathedral and the Bizarre

#109

Post by BlueTemplar »

saagarjha wrote: post #94
bsder wrote: post #61

Earlier quoted context omitted.

> I have like 30 projects on github, I don't expect any of them to really go anywhere, it's just where I put stuff when I'm playing with a new concept. Then why put it there? git works fine without github. And by putting your projects on github all you did was add to the baseline white noise that people have to filter through.

Where should I put it? Or are you suggesting not sharing it at all?

My view is different from the person you're answering to, but - on an actual open-source based website that hasn't been bought by a closed-source transnational corporation ?

Re: The Cathedral and the Bizarre

#110

Post by BlueTemplar »

colin_mccabe wrote: post #92

The harsh reality is that most software, no matter whether it's open source or closed, never makes it big. Take a stroll through an app store on iPhone-- or better yet, on Android! just to see how many apps there. You'll find hundreds of flashlight apps, calendar apps, and so on. Buggy, abandoned software is hardly unique to open source. Then there is the argument that "open source isn't innovative." This seems like…

Tensorflow is backed by Google, an advertising transnational corporation.
Post reply on HN