3 ms·

> However, it's crucial that we have multiple reliable launch and crew providers to ensure the safety and sustainability of space missions. The keyword here is

> However, it's crucial that we have multiple reliable launch and crew providers to ensure the safety and sustainability of space missions. The keyword here is reliable and I would add the word competitive. Having an expensive, late, unreliable provider may in fact be a net negative. I think Starliner in it's current form isn't helping the industry. I hope they get their act together, or that we fund a reliable and competitive alternative to SpaceX.

firesteelrain 2y ago
IMHO, Starliner is too big of an MVP
- windexh8er 2y ago
  That's because Starliner isn't an MVP. It's a vehicle designed to transport humans. You don't send humans to space in an MVP.
  - firesteelrain 2y ago
    To be clear, you can have multiple and iterative MVPs
    - dotancohen 2y ago
      Then at least one of them isn't M or at least one of them isn't V.
      - firesteelrain 2y ago
        Min Viable .. sure you can do that. I guess I am thinking of a different definition than you of viable
        dotancohen 2y ago
        I'm saying that if you do multiple MVPs, then at least one of them isn't Minimum or one of them isn't Viable.
        firesteelrain 2y ago
        A lot of people get that wrong. It’s certainly not how we treat MVPs in my particular industry. “ Yet what is often missed is that a minimum viable product isn’t merely a stripped down version of a prototype. It is a method to test assumptions and that’s something very different. A single product often has multiple MVPs, because any product development effort is based on multiple assumptions.” Article on the topic: https://greg-satell.medium.com/heres-what-most-people-get-wrong-about-minimum-viable-products-9de53ec6e6ea https://greg-satell.medium.com/heres-what-most-people-get-wr...
        dotancohen 2y ago
        > It is a method to test assumptions and that’s something very different. Products shouldn't test assumptions, that's what the PoC is for. Proving (or disproving) a concept.
        firesteelrain 2y ago
        I will disagree. You are building a product and you need to know if you are building the right thing. So you build it using a set of MVPs. Another way to call it is Objectives
        windexh8er 2y ago
        You still get it very wrong. MVP focuses on developing a product with just enough features to satisfy early users and provide feedback for future iterations. This means that the full vision of the product is sacrificed in favor of speed and minimalism. Key word: sacrificed. You don't sacrifice in a system humans are reliant on to live. You can find this described on Wikipedia in a similar manner [0]. Compare this to a Waterfall approach - an approach that has been used in developing rigor in critical systems for decades. Waterfall emphasizes a complete and well-documented design upfront, ensuring that the final product aligns with the original vision and objectives. The end result is a fully-featured product, even if (and it will) take longer to develop. Again... Wikipedia [1]. So, no. You don't build a mission critical system by stacking MVPs like Lego block on top of each other and then calling them "objectives". It's clear you've never built or been involved with building systems that are classified as "Safety of Life Critical System". Feel free to go review some relevant standards (e.g., ISO 26262, IEC 61508, DO-178C) and then feel free to re-justify how MVPs could be used for space vehicles that transport humans. [0] https://en.wikipedia.org/wiki/Minimum_viable_product https://en.wikipedia.org/wiki/Minimum_viable_product [1] https://en.wikipedia.org/wiki/Waterfall_model https://en.wikipedia.org/wiki/Waterfall_model
  - flamedoge 2y ago
    so whats it good for. sending monkeys?
  - ajcp 2y ago
    > That's because Starliner isn't an MVP. > You don't send humans to space in an MVP. You are technically correct given that minimum viability for Starliner is sending humans to space AND returning them to earth. At this point I'm not sure anyone at NASA would agree that Starliner should have been used to send humans to space even.
  - chrsig 2y ago
    I mean, I wish that didn't need to be said, but after the whole submarine-controlled-by-an-xbox-going-to-the-titantic thing, who am I to say what bucket of bolts people might jump into entirely on their own accord
  - michaelcampbell 2y ago
    So what do you call and MVP of transporting humans?
    - mulmen 2y ago
      The ability to return, alive.
      - michaelcampbell 2y ago
        So there IS an MVP of transporting humans. (Which is the point.)
- nordsieck 2y ago
  > IMHO, Starliner is too big of an MVP Maybe. But it's a lot simpler than NASA's previous vehicle, the Shuttle. If you read the selection statement[1], it seems clear in retrospect that NASA put too much weight on Boeing's Shuttle experience (via Rockwell), and not enough emphasis on SpaceX's Dragon 1 experience. But I think, at the time, it was difficult to know which factor was more important. --- 1. https://www.nasa.gov/wp-content/uploads/2019/08/cctcap-source-selection-statement-5083.pdf?emrc=8a4806 https://www.nasa.gov/wp-content/uploads/2019/08/cctcap-sourc...
- phkahler 2y ago
  >> IMHO, Starliner is too big of an MVP Keep in mind that Starliner is NOT the entire launch system. It is only the crew capsule. It rides on top of an existing rocket. The same is true of SpaceX Dragon which rides on top of Falcon 9 that already existed. To your point, Starliner could have started as cargo-only to prove out as much as possible. That's what Dragon did.
  - whatshisface 2y ago
    A cargo-only starliner still would have suffered the same problem and had the same risk of ramming the station.
    - jfoster 2y ago
      Seems like it would be best to test approaching an imaginary or decoy space station when proving a new vehicle.
      - firesteelrain 2y ago
        Or before trying to do it all including a bellyflop
      - dotancohen 2y ago
        That test was scraped as a cost saving measure. 2016 or 2018 if I recall correctly, after it had been delayed several times.
option 2y ago
Yes, we need multiple providers and the proper way to do that is not by bailing out poor designs by incumbents. Instead, we should be setting lucrative incentives for new entrants.

firesteelrain 2y ago

IMHO, Starliner is too big of an MVP

windexh8er 2y ago
That's because Starliner isn't an MVP. It's a vehicle designed to transport humans. You don't send humans to space in an MVP.
- firesteelrain 2y ago
  To be clear, you can have multiple and iterative MVPs
  - dotancohen 2y ago
    Then at least one of them isn't M or at least one of them isn't V.
    - firesteelrain 2y ago
      Min Viable .. sure you can do that. I guess I am thinking of a different definition than you of viable
      - dotancohen 2y ago
        I'm saying that if you do multiple MVPs, then at least one of them isn't Minimum or one of them isn't Viable.
        firesteelrain 2y ago
        A lot of people get that wrong. It’s certainly not how we treat MVPs in my particular industry. “ Yet what is often missed is that a minimum viable product isn’t merely a stripped down version of a prototype. It is a method to test assumptions and that’s something very different. A single product often has multiple MVPs, because any product development effort is based on multiple assumptions.” Article on the topic: https://greg-satell.medium.com/heres-what-most-people-get-wrong-about-minimum-viable-products-9de53ec6e6ea https://greg-satell.medium.com/heres-what-most-people-get-wr...
        dotancohen 2y ago
        > It is a method to test assumptions and that’s something very different. Products shouldn't test assumptions, that's what the PoC is for. Proving (or disproving) a concept.
        firesteelrain 2y ago
        I will disagree. You are building a product and you need to know if you are building the right thing. So you build it using a set of MVPs. Another way to call it is Objectives
        windexh8er 2y ago
        You still get it very wrong. MVP focuses on developing a product with just enough features to satisfy early users and provide feedback for future iterations. This means that the full vision of the product is sacrificed in favor of speed and minimalism. Key word: sacrificed. You don't sacrifice in a system humans are reliant on to live. You can find this described on Wikipedia in a similar manner [0]. Compare this to a Waterfall approach - an approach that has been used in developing rigor in critical systems for decades. Waterfall emphasizes a complete and well-documented design upfront, ensuring that the final product aligns with the original vision and objectives. The end result is a fully-featured product, even if (and it will) take longer to develop. Again... Wikipedia [1]. So, no. You don't build a mission critical system by stacking MVPs like Lego block on top of each other and then calling them "objectives". It's clear you've never built or been involved with building systems that are classified as "Safety of Life Critical System". Feel free to go review some relevant standards (e.g., ISO 26262, IEC 61508, DO-178C) and then feel free to re-justify how MVPs could be used for space vehicles that transport humans. [0] https://en.wikipedia.org/wiki/Minimum_viable_product https://en.wikipedia.org/wiki/Minimum_viable_product [1] https://en.wikipedia.org/wiki/Waterfall_model https://en.wikipedia.org/wiki/Waterfall_model
- flamedoge 2y ago
  so whats it good for. sending monkeys?
- ajcp 2y ago
  > That's because Starliner isn't an MVP. > You don't send humans to space in an MVP. You are technically correct given that minimum viability for Starliner is sending humans to space AND returning them to earth. At this point I'm not sure anyone at NASA would agree that Starliner should have been used to send humans to space even.
- chrsig 2y ago
  I mean, I wish that didn't need to be said, but after the whole submarine-controlled-by-an-xbox-going-to-the-titantic thing, who am I to say what bucket of bolts people might jump into entirely on their own accord
- michaelcampbell 2y ago
  So what do you call and MVP of transporting humans?
  - mulmen 2y ago
    The ability to return, alive.
    - michaelcampbell 2y ago
      So there IS an MVP of transporting humans. (Which is the point.)
nordsieck 2y ago
> IMHO, Starliner is too big of an MVP Maybe. But it's a lot simpler than NASA's previous vehicle, the Shuttle. If you read the selection statement[1], it seems clear in retrospect that NASA put too much weight on Boeing's Shuttle experience (via Rockwell), and not enough emphasis on SpaceX's Dragon 1 experience. But I think, at the time, it was difficult to know which factor was more important. --- 1. https://www.nasa.gov/wp-content/uploads/2019/08/cctcap-source-selection-statement-5083.pdf?emrc=8a4806 https://www.nasa.gov/wp-content/uploads/2019/08/cctcap-sourc...
phkahler 2y ago
>> IMHO, Starliner is too big of an MVP Keep in mind that Starliner is NOT the entire launch system. It is only the crew capsule. It rides on top of an existing rocket. The same is true of SpaceX Dragon which rides on top of Falcon 9 that already existed. To your point, Starliner could have started as cargo-only to prove out as much as possible. That's what Dragon did.
- whatshisface 2y ago
  A cargo-only starliner still would have suffered the same problem and had the same risk of ramming the station.
  - jfoster 2y ago
    Seems like it would be best to test approaching an imaginary or decoy space station when proving a new vehicle.
    - firesteelrain 2y ago
      Or before trying to do it all including a bellyflop
    - dotancohen 2y ago
      That test was scraped as a cost saving measure. 2016 or 2018 if I recall correctly, after it had been delayed several times.

option 2y ago

Yes, we need multiple providers and the proper way to do that is not by bailing out poor designs by incumbents. Instead, we should be setting lucrative incentives for new entrants.