18 ms·
I work in finance. About 5 years ago I started building a FIX library on my spare time, out of curiosity. Over the years it has been countless fin tech start-up
by logotype 5y ago
I work in finance. About 5 years ago I started building a FIX library on my spare time, out of curiosity. Over the years it has been countless fin tech start-ups as well as big companies reaching out to me about the library, suggesting fixes and features. Since then I have long lost the interest in the technology which enables connectivity to financial exchanges to automate trades, but I keep working on the library for the benefit of others and just the joy of creating something. But what’s the actual impact? Enabling companies getting richer? Greed?
https://gitlab.com/logotype/fixparser https://gitlab.com/logotype/fixparser
- CydeWeys 5y agoAre your users compensating you well for this? Sounds like the impact you're having is making fintech serious money, so I hope you're getting some of that.
- logotype 5y ago0 USD :)
- bigyikes 5y agoI don’t think you have any obligation to continue supporting the software, but give yourself more credit! You’re making dozens (hundreds? thousands?) of developers lives significantly better. You’re helping bring the finance industry out of the stone age. You’ve created something useful and valuable, which is no small feat.
- hdudhdhd 5y agoIsn't FIX messaging part of the stone age problem?
- bigyikes 5y agoWell, baby steps and all that :^) At least it’s not a hand-rolled, “enterprise-grade” (read: extremely verbose with unnecessary abstractions) pile of spaghetti. I’m slightly scarred from my brief stint at a large finance firm.
- agumonkey 5y agoI'm trying to find niches where positive feedback loops have a nicer effect on society (right now trying to help courthouse daily duties for instance)
- akudha 5y agoIt would be nice to make a list of such niches. Things that wouldn’t take more than a couple of hundred hours to make (at max) and are tremendously useful, especially to vulnerable populations like old people, people with limited internet connection, physically challenged people etc. These need not be super polished, optimized etc. They just need to work correctly and solve a problem quickly and easily
- agumonkey 5y agoold people are a special demographics (I encounter a few of them regularly, for IT reasons or not) the obvious fields in the list: - education - healthcare - justice someone on HN showed an app for people to find good workers for home rework, this kind of stuff could be tremendously useful, and more generally the organization layer of society is very lacking. This also match the issues at the courthouse, basically making people well informed, and communicating issues/info smoothly (instead of waiting days to get a fake answer by someone shady)
- akudha 5y agoPhysically challenged people are another group that could use a lot of help. One of my friends has cerebral palsy, until I met her, I had zero idea about the problems they face. These groups of people aren’t served because there is not much money in it :(
- agumonkey 5y agoI'd gladly help people with those issues. I'm gonna seek them (no idea why I never thought about that). The reason I put elders aside is that their motivations and emotions are very different from younger people. There are lot of fear, loneliness driving their need for help, and often they don't really want to own the knowledge, it's more a little bit of emotional support; which is cool, but not as motivating as, say, a younger handicapped person, who has more skin in the game.
- alien_ 5y agoI'd probably offer some sort of support/consultancy services and prioritize the work of the paying users, making it clear when filing github issues and pull requests. This could easily sustain you in a passive income way so you can spend most of your time working on things that matter more to humanity.
- eloff 5y agoYeah, I think especially in finance companies would be willing to pay for support and feature development. There has to be a way to monetize the project and make it so at least it pays for your time. There's nothing wrong with a labor of love, but the companies using your software can afford to pay for your time.
- Asymmetryk 5y agoI am probably not actually being realistic in terms of economic results and ROI, but the FIX library GP seems well established - making it documentation that's the case I would like to put forward : any code in a critical path, and what better than a financial exchange subsystem, can be purchased from any number of vendors with support, but how many have genuinely caring, intelligent, thoughtful and usable beyond the immediate remit, thorough and readable documentation? Cloth letter binder and loose leaf format with theoretical introduction, theory of operation, enumerated common implementation errors for examples together with laying out conventions for the payload / XML DTD - the docs you can write tests and tools against without the code.... I have signed off on a pretty penny or two to obtain docs like this, and hundreds of dollars per copy is totally reasonable. I'd thin film laminate select chapters and affix a dry wipe pen inside (how much would it be to custom some from All express?) and by the sounds of this situation I think you could see a standard adoption and satisfying return. Boxes, ring bind, shrink wrap and a 2nd color printed sparingly add up to something that management likes a lot, usually. Professional product support bridges the economicd to enterprise fees, imo..
- kilroy123 5y agoBut then it's a job and he/she probably won't like it anymore.
- satellite2 5y agoI've been a user of your project and it's fantastic. I think you're being too hard on yourself. The proliferation of marketplaces and assets to trade (and all type of projects, automated trading, market making, etc... that your project enabled) can been seen as a net positive, as a win win for all parties involved and more. From a microeconomic perspective, your library enabled access to startup using js and with relatively simple tech stacks access to a tool that would otherwise be relatively complicated and time consuming and for which they would not necessary have the resouces. By doing so you lowered the cost of entry and allowed new entrants to challenge the status quo. And from a macroeconomic perspective I think it's one of the reason access to capital has been simplified and has lowered in cost, it's one of the reason for the low cost of mortgages and the relative resilience of economies during covid (central banks interventions are useless without bank and other financial intermediaries participation, it would be like pushing on a string). Don't be fooled by the contemporary contempt against finance. It's still the most important reason that explain the constant improvmements in prosperity and peace of recent years, and the best thing one can do is to make it less arcane and open it to more people. One of the achievements of your tool. You can be proud of your work.
- nicoburns 5y ago> Don't be fooled by the contemporary contempt against finance. It's still the most important reason that explain the constant improvmements in prosperity and peace of recent years I think the contempt comes from the fact that financial instruments have a tendency to concentrate those benefits in small proportion of the population. Low-cost mortgages are an excellent example: these have enabled wealth gain for those who were able to get them at a certain time. But that gain has come almost entirely at the expense of those who weren't able to (or simply didn't) get them, who now face greatly increased rent and house prices.
- lazide 5y agoIt isn’t financial instruments doing that - it’s people. Be thankful that it allows this to happen somewhat transparently and peacefully, rather than at the hands of pinkertons, criminals, or the military - which is how historically it has happened.
- boulos 5y agoHuh! You left out the most funky part: you wrote a FIX parser in Typescript. What are people using it for? To make web apps and just not have to roll their own wrappers around FIX/FIX-derived data? (That is, if all back ends are used to speaking FIX, it’s nicer to have the front end and web serving tools also just speak it?).
- true_religion 5y agoPeople use nodejs in finance backends too. Its async operators make it easy to munge different data streams together and fan in/out results.
- vmception 5y agooh this is something to say louder for the people in the back: there is no reason that finance and computational finance gravitated around python there is nothing inherent about that language aside from people not having made some libraries in other languages javascript especially nodejs is fine for most people's finance cases
- nuclearnice3 5y agoIt's true there is nothing inherent. Because all these non-statically typed garbage collected languages with Algol-like syntax are basically the same language. Given that it makes sense that the decision is made based on libraries, community, inertia and whim.
- lmm 5y ago> Huh! You left out the most funky part: you wrote a FIX parser in Typescript. What's odd about that? Typescript is a boring middle-of-the-road language from the last ~10 years, it's a pretty reasonable default language (although structural typing is a bad default if you ask me).
- high_byte 5y agoif you want to erase billions and stop the richness just remove the library, or better yet add a typo :D
- jjeaff 5y agoThere is a common refrain in charitable organizations that are focused on helping the poor: "Sometimes you have to feed the greedy to get to the truly needy." In other words, the greedy will always take advantage, but that doesn't diminish the help you give to the needy. There is simply no way to totally filter for the greedy. This doesn't really apply directly to your situation, but at least for every greedy company that benefits from free products like yours, there are a lot of jobs created and benefit to society as a whole when these companies create useful products.
- cortesoft 5y agoI do find this aspect of human morality interesting... people get REALLY upset at freeloaders, to a degree that is not proportionate to the cost they add. We choose to spend a ton of resources to prevent freeloading, even when the amount we spend to stop them is more than we actually lose to the freeloaders. I guess it kinda makes sense as social creatures, but it can be frustrating at times when you are trying to help people.
- petschge 5y ago> We choose to spend a ton of resources to prevent freeloading, even when the amount we spend to stop them is more than we actually lose to the freeloaders. Is it also more than what we would lose to freeloaders if we did nothing against them? If yes, then it's indeed not rational. But if no, then it's just indicative that what we are doing is working.
- cortesoft 5y agoYeah, that is part of why I said it was interesting… it is hard to figure out how important that behavior is for social stability.
- Aeolun 5y agoI imagine if we stopped doing those things the number of freeloaders would increase. So the net benefit is negative now, but it’s still better than the cost of not having the measures at all. Also, yeah, it shouldn’t be easy to freeload. Make them work for it!
- mech4bg 5y agoThis is amazing. I now work in consumer SAAS web development, but I worked for a long time in finance and for 7 years of that exclusively focused on FIX, writing a real-time, highly performant market interface. I would have _loved_ to have had something like this while I was working on it. Have you thought of commercializing it? Do you get to use it in your day-to-day work?
- qmmmur 5y agoAre you seriously asking that question about people who want to automate trading in the stock market?