7 ms·

Six months ago, I started developing a Flight Alerting service for QPX Express API. I was always skeptical as to how long would they keep this API running, sin

Six months ago, I started developing a Flight Alerting service for QPX Express API. I was always skeptical as to how long would they keep this API running, since it was a byproduct of a previous acquisition. But I said to myself, wishfully thinking: Why would they shut down this and lose their clients trust in developing with their APIs? Sadly, I was again proven wrong.

FeepingCreature 9y ago
If a client hasn't learned by now, they're not gonna learn from this either. For niche apis Google has no trust left to burn, because all trust in its products is either already burnt or inflammable.
- helloguillecl 9y ago
  Sadly true
- stefano 9y ago
  And yet, there are still a lot of companies using and relying on Google APIs and cloud services. I don't understand why there is still so much trust in them after they've shown multiple times they will terminate services with little warning. And no, I don't consider 1 year a long enough warning for something that has no clear drop-in substitute. If you're using a Google service, you must be prepared for when they'll pull it from under your feet. And if you need to prepare for that, why use that service to begin with if an alternative exists?
  - cortesoft 9y ago
    Isn't that true of ANY company relying on some service from some other company? What makes you so sure that this alternative service is more likely to stay around than one of Googles? The reason we think that Google EOLs so many more APIs than other companies is because Google itself is still around. Some other company's API wont be EOLd, instead they will just go out of business (or they will be bought by someone who will shut down the service).
    - bduerst 9y ago
      Also known as survivorship bias. You don't see the failed companies with dropped APIs because the company disappeared.
    - bitL 9y ago
      Are you going to invest into a company that has such an API its bread-and-butter, forced by finances to continuously improve it, or into a company that has it as a shiny new toy and throws it into basket when something else becomes more shiny?
      - s17n 9y ago
        You mean a company like ITA, for example?
  - PeterisP 9y ago
    If you're using any service that's absolutely required for business, then you must be prepared for when they'll pull it from your feet or increase the pricing to eat all your profit margin. Of course you need to prepare for that, that's just basic part of running any business, software or not. The answer to "why use that service to begin with if an alternative exists?" is likely based on cost or quality, but if it's required for your business and you can't switch to an alternative within 1 year, then you obviously knew that you should have had started looking for an alternative long ago and start preparations for switching to an alternative, from day one or, more precisely, from the day you decided that this business is worth running at all. The only excuse not to do that if it's a commodity product with clear drop-in substitutes, otherwise you're just sloppy with running a business and/or knowingly taking on risks that your business may suddenly stop - and not because of the API going away (the risk was predictable) but because of your own unpreparedness. It's the same with any other services. What do you do if your landlord dumps you or the building burns down or goes down in a flood or hurricane or whatever? If your need for facilities is a commodity in a liquid market, then you'll be able to replace them, but if not, if you need something that's not readily available to avoid huge losses then it's your (assuming executive positions) duty to prepare alternatives in time and/or get appropriate insurance. What do you do if your service for accepting credit cards stops working or dumps you for whatever reason? If you're a reasonable business, you have a second, independent contract with all the deals already arranged that you can switch to in a day if not immediately. The same goes for any non-commodity that you need. If a supplier dumping you means just that you have to drive to a store and buy the same thing from someone else more expensively, then you can skip worrying, but if not, then it's on you to safeguard your business with alternatives instead of blaming suppliers for being unreliable. In some cases, the suppliers will actively exploit knowing that you don't have prepared alternatives - if they've got you by the balls, they can squeeze them and get the price and conditions that they want; and if/when you get serious alternatives, then you can get a whole new range of discounts, that's business as usual.
- Filligree 9y ago
  Nit: "inflammable" means "can be set of fire". So does "flammable". Yes, I know. You want "not flammable", or if you want to be cute you could try "unflammable".
  - mkartic 9y ago
    Looks like the post means 'either already burnt or will be burnt easily'? So what exactly is your nitpick here?
    - Dylan16807 9y ago
      That doesn't make much sense. Why would the easily-burnt trust still be intact at this point? It's a really weird comment if it wasn't supposed to mean "already burnt, or unable to burn".
    - rhizome 9y ago
      I read it to mean 'already burnt or can't be burnt.' I'm aware of the definitions of 'inflammable,' but it just made the most sense in the context to interpret it that way.
    - FeepingCreature 9y ago
      I can't edit it anymore, but I did mean "unburnable". I realize (now) that "inflammable" comes from "inflame"/"enflame", but "in-"/"im-" has lots of use as a negation. "Invisible", "indivisible", "inconceivable", etc.
  - krisdol 9y ago
    I think inflammable (or flammable) applies. I think the user is saying the trust is likely to go up in flames.
  - _ah 9y ago
    Interestingly, I read this the other way: "...all trust is burnt, or about to be burnt."
  - JohnHaugeland 9y ago
- jpalomaki 9y ago
  I was looking into this space some time ago. Got a feeling that this offering was quite unique, I don't think other vendors are offering this information with pay-as-you-go arrangements suitable for small projects. (But I might be wrong with this) Edit: Looks like other comment chain somebody has mentioned an alternative: https://www.fareportallabs.com/Home/DownloadDocs#0 https://www.fareportallabs.com/Home/DownloadDocs#0
tanilama 9y ago
If it is free, and no ads, it is probably not going to stick long
- nathancahill 9y ago
  It's not free, it's expensive.
- specialp 9y ago
  Correction: If it is not resulting in you transferring more data from yourself or others to Google to profit off of, it is not going to stick around long. Google doesn't seem interested in running paid businesses besides advertising no matter how much you are paying them.
  - mandelbrotwurst 9y ago
    Umm...Google Cloud Platform?
  - s17n 9y ago
    Google is definitely interested in "paid businesses". What they aren't interested in is businesses under a billion* dollars. *Made up and likely exaggerated number
    - htrp 9y ago
      Actually true... Google had amazing robot grasper arms once upon a time. These were much better than anything that was in industry. The team got killed because it wasn't a "toothbrush test" business
      - mgkimsal 9y ago
        i'm curious why google (and other large companies like that) might not try to sell those sorts of tech internal-startups to other industry players. possibly a lot of hassle, no doubt, but... tens of thousands of man hours and research just get shelved because "toothbrush test" (then... why were some of these obviously-not-toothbrush ideas being pursued in the first place) and no one else gets to benefit?
un_montagnard 9y ago
I was going to release a side project based on QPX Express API that allows comparing prices for a flight in different countries. Guess I won't bother fixing the last details and will just add it to my list of never finished things.

FeepingCreature 9y ago

If a client hasn't learned by now, they're not gonna learn from this either. For niche apis Google has no trust left to burn, because all trust in its products is either already burnt or inflammable.

helloguillecl 9y ago
Sadly true
stefano 9y ago
And yet, there are still a lot of companies using and relying on Google APIs and cloud services. I don't understand why there is still so much trust in them after they've shown multiple times they will terminate services with little warning. And no, I don't consider 1 year a long enough warning for something that has no clear drop-in substitute. If you're using a Google service, you must be prepared for when they'll pull it from under your feet. And if you need to prepare for that, why use that service to begin with if an alternative exists?
- cortesoft 9y ago
  Isn't that true of ANY company relying on some service from some other company? What makes you so sure that this alternative service is more likely to stay around than one of Googles? The reason we think that Google EOLs so many more APIs than other companies is because Google itself is still around. Some other company's API wont be EOLd, instead they will just go out of business (or they will be bought by someone who will shut down the service).
  - bduerst 9y ago
    Also known as survivorship bias. You don't see the failed companies with dropped APIs because the company disappeared.
  - bitL 9y ago
    Are you going to invest into a company that has such an API its bread-and-butter, forced by finances to continuously improve it, or into a company that has it as a shiny new toy and throws it into basket when something else becomes more shiny?
    - s17n 9y ago
      You mean a company like ITA, for example?
- PeterisP 9y ago
  If you're using any service that's absolutely required for business, then you must be prepared for when they'll pull it from your feet or increase the pricing to eat all your profit margin. Of course you need to prepare for that, that's just basic part of running any business, software or not. The answer to "why use that service to begin with if an alternative exists?" is likely based on cost or quality, but if it's required for your business and you can't switch to an alternative within 1 year, then you obviously knew that you should have had started looking for an alternative long ago and start preparations for switching to an alternative, from day one or, more precisely, from the day you decided that this business is worth running at all. The only excuse not to do that if it's a commodity product with clear drop-in substitutes, otherwise you're just sloppy with running a business and/or knowingly taking on risks that your business may suddenly stop - and not because of the API going away (the risk was predictable) but because of your own unpreparedness. It's the same with any other services. What do you do if your landlord dumps you or the building burns down or goes down in a flood or hurricane or whatever? If your need for facilities is a commodity in a liquid market, then you'll be able to replace them, but if not, if you need something that's not readily available to avoid huge losses then it's your (assuming executive positions) duty to prepare alternatives in time and/or get appropriate insurance. What do you do if your service for accepting credit cards stops working or dumps you for whatever reason? If you're a reasonable business, you have a second, independent contract with all the deals already arranged that you can switch to in a day if not immediately. The same goes for any non-commodity that you need. If a supplier dumping you means just that you have to drive to a store and buy the same thing from someone else more expensively, then you can skip worrying, but if not, then it's on you to safeguard your business with alternatives instead of blaming suppliers for being unreliable. In some cases, the suppliers will actively exploit knowing that you don't have prepared alternatives - if they've got you by the balls, they can squeeze them and get the price and conditions that they want; and if/when you get serious alternatives, then you can get a whole new range of discounts, that's business as usual.
Filligree 9y ago
Nit: "inflammable" means "can be set of fire". So does "flammable". Yes, I know. You want "not flammable", or if you want to be cute you could try "unflammable".
- mkartic 9y ago
  Looks like the post means 'either already burnt or will be burnt easily'? So what exactly is your nitpick here?
  - Dylan16807 9y ago
    That doesn't make much sense. Why would the easily-burnt trust still be intact at this point? It's a really weird comment if it wasn't supposed to mean "already burnt, or unable to burn".
  - rhizome 9y ago
    I read it to mean 'already burnt or can't be burnt.' I'm aware of the definitions of 'inflammable,' but it just made the most sense in the context to interpret it that way.
  - FeepingCreature 9y ago
    I can't edit it anymore, but I did mean "unburnable". I realize (now) that "inflammable" comes from "inflame"/"enflame", but "in-"/"im-" has lots of use as a negation. "Invisible", "indivisible", "inconceivable", etc.
- krisdol 9y ago
  I think inflammable (or flammable) applies. I think the user is saying the trust is likely to go up in flames.
- _ah 9y ago
  Interestingly, I read this the other way: "...all trust is burnt, or about to be burnt."
- JohnHaugeland 9y ago
jpalomaki 9y ago
I was looking into this space some time ago. Got a feeling that this offering was quite unique, I don't think other vendors are offering this information with pay-as-you-go arrangements suitable for small projects. (But I might be wrong with this) Edit: Looks like other comment chain somebody has mentioned an alternative: https://www.fareportallabs.com/Home/DownloadDocs#0 https://www.fareportallabs.com/Home/DownloadDocs#0

tanilama 9y ago

If it is free, and no ads, it is probably not going to stick long

nathancahill 9y ago
It's not free, it's expensive.
specialp 9y ago
Correction: If it is not resulting in you transferring more data from yourself or others to Google to profit off of, it is not going to stick around long. Google doesn't seem interested in running paid businesses besides advertising no matter how much you are paying them.
- mandelbrotwurst 9y ago
  Umm...Google Cloud Platform?
- s17n 9y ago
  Google is definitely interested in "paid businesses". What they aren't interested in is businesses under a billion* dollars. *Made up and likely exaggerated number
  - htrp 9y ago
    Actually true... Google had amazing robot grasper arms once upon a time. These were much better than anything that was in industry. The team got killed because it wasn't a "toothbrush test" business
    - mgkimsal 9y ago
      i'm curious why google (and other large companies like that) might not try to sell those sorts of tech internal-startups to other industry players. possibly a lot of hassle, no doubt, but... tens of thousands of man hours and research just get shelved because "toothbrush test" (then... why were some of these obviously-not-toothbrush ideas being pursued in the first place) and no one else gets to benefit?

un_montagnard 9y ago

I was going to release a side project based on QPX Express API that allows comparing prices for a flight in different countries. Guess I won't bother fixing the last details and will just add it to my list of never finished things.