Yeah I built one of these for .NET specifically for this, with DB-specific drivers to extend the API, especially for Postgres. It just tries to be a dead simple representation of the query, then provides a command API to do mapping and execution, some unit of work support. Really happy with it so far, especially compared to an ORM.
Every cool new revolutionary, game changing, disruptive tech I can think of has had some kind of “killer app” that lead the way to widespread pubic awareness. Peer to peer had Napster, AJAX had Google Maps, microservices had services of large scale tech giants, and even blockchain originally had Bitcoin. In each case people were doing some cool new thing and we approached the technology trying to figure out how it was accomplished. I am not aware of a single such example for Web3 nor has anyone I’ve asked been able to name one. The killer Web3 apps are at best hypothetical imaginings of what someone could build with it someday, and even that take is quite generous. It is a technical buzzword in search of funding which is not a proven solution for anything at all.
If I wanted to send you $50 worth of assets, literally right now, what alternatives do we have apart from crypto? Certainly nothing that does such a good job of preserving our relative privacy to each other. I don't want to, but it is still an interesting question.
There are capabilities here that have never existed before, it seems likely that something radical will happen in time. There is already a lot of money sloshing around, and it takes a while for innovations to sink in.
A regular bank transfer? I know the US doesn't really do that kind of thing, but here in Europe I can send money to another bank account within 10 seconds.[1] And it doesn't even cost me or you anything and all I need is your IBAN.
Here is a Monero address. Things sent there ends up with me. I advise against sending anything to it. Are you going to post your IBAN and account details?
I don't actually know what my IBAN is, but assuming it is my account number I'm not comfortable posting it here like I just did for a crypto address. This technology is something other than what I've had access to before.
Have you tried sending money across the world in 10 seconds, to somebody outside of the EU? Most answers to this are PayPal, Wise, or another multibillion dollar private enterprise.
At least I know that those 5€ only change on a term of months, or with inflation.
Last time I looked at ETH and BTC I spent a good 30mins trying to figure out how much the transaction would actually cost me and if it depended if I clicked GO at an opportune moment. And yes, I admit that this might have only been a difference of cents, but I wouldn't know that beforehand.
You say assets, but cryptocurrencies are not assets. They're funny little monopoly money bills that have no value in the outside world. If you send me $50 in ETH or any large, largely used coin, you're going to pay the same amount in fees. OH and by the time it arrives, it might be worth $45, or $55, you'll never know. Or you could send me a "stable" coin and hope you don't get rug pulled in the middle. Then I have to grab it and find a sucker that wants to buy it from me at exact value (either exchanges, which are going to find bigger suckers than them, or individuals in the street, which don't exist.). But we both know it's not going to happen because why would the sucker on the street buy from me instead of a "trusted" exchange, and why would the "trusted" exchange just take it from me at market rates, instead of charging fees and lowering the exchange rate to make a profit on it.
In the real world, if you have an actual asset, you physically ship it, and it costs money, but I have something in my hands. In cryptocurrency fantasyland, I can buy drugs with it. Or shitty NFTs.
You seem stuck in 2018. No one would use ETH to transfer $50 worth of money unless they already have that and the recipient wants ETH. But if you actually care about fees and speed you would not use ETH.
(I wont shill any specific coin but there are plenty with Tx speed in the seconds and fractions of a penny fees.)
And if you are actually worried about the price movements in the 3 seconds you could simply use a stable coin but realistically its only in your mind since "holding" a volatile asset for a few seconds is way less risky that holding a non-volatile asset for long time (which you likely do all day every day and dont even care about).
Needless to say that DEXes exist as well so you can exchange without using a "trusted" exchange.
And CBDCs aka "gov stable coins" will come as well at which point the risk of holding it is exactly the same as holding paper money and less risk that holding fiat in your bank account because CBDCs would be a balance direly by the central bank so you avoid the traditional bank which could go bankrupt.
Just to be clear I dont think using crypt as a replacement for cash is a proper use case.
Buying coffee with crypto and all that stuff is rather silly at least today.
I think crypto as a means for payment has its real use case in international transactions.
There are millions of people who send money back home from all around the world and no traditional banking dependent service can even remotely compete with crypto. You can no use all the fast local apps that do not actually move any money just change balances between its own users.
Its only a matter of time before companies will use crypto for this as well if they deal with international payments. Crypto in this case will just be the bridge currency and its value or whether it goes up or down does not matter. I might send CAD and you get USD the exchange is done automatically from CAD to crypto to USD. You dont need a bank who takes a fee or gives a bad rate because this exchange can run over public DEX where everyone get the best rate aka the actual market rate.
Nothing outside of crypto can do this. Every other solution has a middleman who sucks out a profit and make the system worse for everyone + is a risk especially if that middlemen is in another jurisdiction.
Sending any amount of ETH right now is going to cost you more like $3-6 of fees, not $50. This has been standard for a while. Other than a spike yesterday with the FTX fallout, fees for sending ETH have been in the $1-10 range for months now
And are you sending and receiving often 50$ worth of assets? It's one thing to say it's possible (and everybody will agree) and another thing is that sending it this way is "better" - for whatever definition of better. But back to my question: are you using this mode yourself? Often?
Cash. In an envelope. To an anonymous PO box. Or text a cash card or amazon voucher code to a burner number. Use a drop gang or mule. Find a Hawala broker. Etc. etc.
I can't tell at this point whether you are trolling, or one of the HN hivemind that actually believes sending cash in an envelope is somehow faster and more secure than using a blockchain and cryptography for payments.
The question was about sending assets and preserving privacy, not about speed or security. Most crypto currencies are the very opposite of privacy preserving, unless you really think that having all your transactions on a publicly available ledger in a form that is trivial to link to your real identity is a good way of preserving anonymity? Yes I know Monero exists. I also know that vulnerabilities in its privacy model pop up every couple of months and you still have to convert it back into something you can use to buy useful stuff and pay your taxes. Doing this means you're going to probably have to go through an exchange that adheres to stricter KYC rules than the average bank. So again, not exactly good for preserving privacy. Obviously as assets they have massive disadvantages due to their volatility. And given the frequency of hacks and people losing their keys due to general idiocy I wouldn't exactly call most crypto "secure". I'd actually say that the irreversibility of transactions makes it insecure by design.
Personally if I want to send money quickly and securely, be it within the UK or internationally, I use the banking system. It's cheaper, its faster, its more secure and if someone makes a mistake it is easily reversible. I even send a monthly remittance to Madagascar for free, and its not exactly known for its advanced banking system what with it being the 4th poorest country in the world. Yes it takes 2 to 4 days to get there, but I can pay anyone within the EU instantly or if its an American there's always paypal.
I'm 90% sure at this point that the main reason that cryptocurrencies became so popular (apart from the opportunities for massive fraud and bilking of the gullible) is because the USA has a banking system that has barely evolved since the 1980s. I mean they were still flying cheques around in 2004 and I still have American businesses trying to pay me by cheque on a regular basis in 2022. When you consider it through that lens it's no wonder Bitcoin sounds like a good idea.
Monero, zcash, aztec, tornado cash all work pretty well. Could you link to Monero's current vulnerabilities? There have been some known and theoretical issues with the privacy of Monero and Tornado Cash but it isn't a monthly occurrence. How are any of these protocols less secure than sending dollar bills in an envelope?
Sending money with a UK bank or PayPal is the opposite of privacy preserving. The convenience of it is fine for UK to UK and EU banks, not great for private, secure, instant and international payments.
I'm not a fan of crypto, but when I read things like this, I have mixed feelings. Porn showing homosexual acts is illegal in some countries. Abortion is becoming illegal in many parts of the US.
It's a conflict that isn't going to be ended any time soon: social control is used for good and for evil. We celebrate social disorder in Tiananmen Square and fear it at the U.S. Capitol. It's not like we can pick one side, order or disorder, and feel completely safe with our choice.
I don't see how I would buy drugs or conduct terrorism on HN. And while avoiding tax on $50 might be a win, it is a boring part of the scenario.
I don't want to post PII info publicly from this account. But, in practice (ie, we're long past theory) if I want to accept payment as roenxi I can. And if I use something like Monero, it'll be extremely anonymous. There'll be a use for that sort of tech to someone and for legitimate purposes.
Not a specific piece of software, but a characteristic of a limited percentage of anonymous internal systems--one of the most beautiful things to witness in the realm of design is a system undergoing catastrophic stress when the designers anticipated and planned for such events, and in usually a very short period of time you see the results of extensive planning spool out, design features hidden from view and unappreciated until this moment, kick in and recover/compensate in ways that feel almost magical.
For obvious reasons it is much more common to see this level of design in physical, life-critical systems like aerospace or automotive technology, but you do see it sometimes in software. Well designed services that under heavy load, various kinds of infrastructure failure, attack, or other kinds of scenarios well outside the bounds of normal expected operations intelligently compensate while signaling alerts with precise, useful information, and attempt whatever kind of recovery is possible.
This is hard to anticipate and often thankless to build in advance. It's always a stressful time when this behavior is visible, but it gives me a feeling of admiration for the perhaps long gone employees who built it.
Mercedes has a system on the S-class that uses the sensors to detect a potential impact (e.g. an rapidly approaching vehicle while the car is stationary) and uses the active suspension to "jump" raising the car a few inches just at the moment of impact, apparently slightly reducing the potential for injury.
Many cars and especially planes & spaceships are have tons of systems like this.
I worked on a system that did things like in response to a primary datasource being offline would switch the queries it used to a different database to include substitute data and pass this back to be used instead until the primary store came back online. Three years after it was put in production this happened on one of the company's biggest and most important days of the year, and our SREs were sitting there calmly trying to solve the issue, ended up waiting until late at night to deploy a fix. It would have been reasonable for this service to just rely on the primary source, but we would have been offline for hours if this little trick hadn't been put in place.
>...a different database to include substitute data and pass this back to be used instead
The thing that trips me up about this is how can you ensure that the substitute data is actually useful. Is the substitute data a copy of the primary data? I guess it all depends on the use case...
Haha, maybe not? Since their car would tend to go underneath, although the footage I've seen always shows the other vehicle hitting from the side with the hood sliding slightly beneath the Mercedes...it does seem from watching that would give other vehicle slightly more space to dissipate the impact energy.
Now if they had only bought a Mercedes as well, they could have chosen the self-breaking option to avoid the collision in the first place.
One that I've always thought is really cool since I heard about it is the base of street signs. Instead of being cemented into the ground, signs are commonly bolted to a base so that they'll break away when hit. Some even have an incline that will throw the sign up and over your car. After learning this, I've started paying attention to signs I pass on walks or while driving and it's crazy how they are everywhere and I never noticed.
Another thing to look into if you find that cool is guard rail design. Modern guard rails have some cool features designed to reduce risks associated with hitting them.
Nortel DMS-10 telephony switch springs to mind. They run in landline phone company switching offices amd date back to the early 1970s. Those things are indestructible in the face of all kinds of disasters.
I suppose it's possible that their auto-play feature is doing it. Most of our Netflix use is on a laptop so if the browser tab isn't closed maybe it keeps streaming the auto-play, although I haven't checked for that.
> Uber charges drivers 25-30 percent without coming close to covering their actual costs
I mean this is it, right? This is the big one. All these "tech" companies that are not actually selling software taking huge cuts off the top line, this can't be sustainable.
You can own the market for a while, but sooner or later someone is going to come along and disrupt you.
Yeah, no. Companies that care about costs absolutely track what they’re spending and where. I’ve seen plenty of merely questionable expenses a fraction of this caught and called out at very large companies.
Indeed, but that's also true for any large organization, whatever their status: public, private, nonprofit.
I witnessed large sums being spent for very dubious reasons in private firms, mainly in order to meet ridiculous demands by executives. Show-off features despite no interest from customers, "visionary product" e.g. implementing a blockchain or "artificial intelligence" that classic algorithms could do as well for much cheaper.
The idea that private entities pay more attention to money than public entities is a prejudice. At least in modern developed democracies.
Financial scandals in private firms tend to be muted in order to preserve the external reputation: hence the impression that public entities are less careful.
It's all the same everywhere, no general rule in that matter.
Both can be true. Companies will question every little nickel and dime certain employees spend and then have big stupid horrible blind spots in other places. Happens all the time. I was a PhD student at Yale and they were extremely touchy about certain kinds of spending, and then would also do stupid stuff like this.
In my experience, they don’t. But it was just a street-name indicating sign that was removed to make way for a sidewalk cut, tossed aside and never re-mounted. Eventually some scrapper took it I bet.
Hell no. Also if this is your expectation you’re absolutely insane if this isn’t automated…also this kind of sounds like QA to me. My presumption is the base functionality requirements are met, I’m looking for potential performance issues, code standards, sound architecture, etc.
I am always amazed at how many repetitive tasks folks want to load up developers with. I think there’s a tendency among developers-turned-managers to see their job as crafting the perfect “program” of processes for their developers to follow, instead just automating these tasks. Like they say, the cobbler’s children have no shoes.
I am afraid that the last few minutes of life will be like a confusing and disoriented nightmare. That has always seemed to me to be the most likely scenario as your brain is shutting down.
If its like psychedelics I don't think it would be. Which i think the initial parts could be similar since they decrease blood flow to the brain, seems the same as dieing? IME there is nothing to fear, fear (emotions in general) is a higher level concept and those seem to be the first thing fall apart. Thoughts about thoughts, self identity, are made up to make living and thinking easier.
my guess is you slowly transition to simple machine with sense inputs and outputs until it all shuts down.
Reading your comment gives me such an unsettling feeling. You nailed it. I think death is the ultimate form of discomfort. Our bodies refuse to let us die peacefully under most circumstances. They fight with every last breath and every last heartbeat, to keep going. And all the usual signals of comfort are gone, with one just left bare before the universe. Think motion sickness + nausea + all the other terrible feelings combined, and you cannot just give in, you just have to endure until you're gone.
Interestingly there are two David Lynch movies (Lost Highway and Mulholland Drive) that can be interpreted as just that: someone's disorienting nightmare as they are dying.
It seems that with these types of changes, I mean with more and more of the workforce working on schedules that don’t overlap, it is going to become more and more important for business software to run autonomously without anyone there to babysit it, but it seems to me we have been moving in precisely the opposite direction for some time now.
Specifically “doing agile” seems to encourage this, I always see lots of supporting infrastructure being skimped on so a new feature can be delivered faster, with some hand wavy plans to fix it in some future sprint.
How on earth is this related to "agile" software/product development?
> I always see lots of supporting infrastructure being skimped on so a new feature can be delivered faster
That's bad management and has nothing to do with agile. Reserve time for bug fixes (or have fire fighting capabilities) and refactoring, include time for tech tickets. ALWAYS. Plan workload jointly with developers, talk to them regularly. It's that easy.
Right, nobody who has a bad experience with agile is doing it right. I am very familiar with this line of argument.
Fact remains though that back when everything was waterfall (proudly) and we shipped physical media to clients, when a bug often meant flying a support engineer out to the customer’s office, software was generally delivered more completed and vetted, and the QA group wasn’t allowed to report to the same org as developers because it was considered a conflict of interest. Now some software companies don’t even have a QA organization, we are inventing new organizational and technical structures to ensure that the developers are perpetually on call to resolve issues, and there is definitely a school of thought that says delivery of a complete product is an anti-pattern because after all every feature is just an experiment that you are testing on your users. But whatever, two different times, two different needs, two different practices, and to my point now we seem to be entering a new post-COVID era where just as waterfall doesn’t scale to a continuous delivery paradigm, “you build it, you run it” as practiced today is going to struggle with an “always on” business which has no “after hours” due to more flexible working schedules.
Are there companies today that run true 24/7? Sure, plenty, but they tend to be larger and have the staff to cover this. A great deal of business software today is still very fragile operationally and has deployment, management, and maintenance tied to the 9-5, M-F business schedule. I think it is going to have to evolve.
> Now some software companies don’t even have a QA organization, we are inventing new organizational and technical structures to ensure that the developers are perpetually on call to resolve issues, and there is definitely a school of thought that says delivery of a complete product is an anti-pattern because after all every feature is just an experiment that you are testing on your users.
That's bad management again imho, not tied to agile. I hope you're not working like that!
We are a small startup and have a ratio of 2 QA per 7 developers. And QA is kind of outside the scrum process (with more QA resources I could have some integrated and working on new features directly, that would be nice, but you know, money and stuff). We actually run automated end-to-end tests via selenium via CI on every new docker image release on a production-like environment. Plus manual QA.
My point wasn't that "agile" wasn't done right, but that management is shit and it wouldn't even make a difference if you used waterfall or agile in this situation.
Also,
> to ensure that the developers are perpetually on call to resolve issues
That's a good way to get rid of your developers lol.
You're points about "always on" business are interesting, but I don't experience it like that. Maybe depends on the country/sector?
Yes, I got that! That's why I used "agile" as well. My first thought was that it might be true what they said because it might be harder to account for said work, but I think in the end it depends on the management style and the priorities. If management and the dev team don't have the right priorities, traditional or agile can't do anything about it. And the technology leader (i.e. a CTO/tech lead etc.) needs to protect devs from pressure and stress of stakeholders and users. You need to foster understanding that developing software takes time and things like proper testing, QA, security, refactoring, updating libs etc. saves you money and allows you to keep the momentum. That's the job of the CTO imho.
Also: if you want to move faster, grow your team (and your structures) and see where time is wasted (we don't have a lot of meetings for devs that are mandatory besides demo, sprint planning, retro and standups and all have a max time) and don't cut corners all the time, they are coming back to bite you most of the time.
If anyone's interested, I can only recommend "Scrum and XP from the trenches"!
/e: sorry for the long text, tl;dr:
have the right priorities to produce good software, agile is not the cause, your shitty management is
