5 ms·

The antivirus isn't great, either. This is a critical issue, because the networking process inevitably results in numerous breaches which frequently affect othe

The antivirus isn't great, either. This is a critical issue, because the networking process inevitably results in numerous breaches which frequently affect other stakeholder systems. Those networking operations critical to a successful eventual deployment of the small child, but the number of bugs experienced is simply inconceivable to users accustomed to good security hygiene.

samwillis 5y ago
The trouble we find is our v2 model, which we had hoped could have made improvements over v1, frequents the same unsavoury places that v1 did initially and continues to be compromised by the same vulnerabilities. It seems later versions are destined to make the same mistakes again. The only comfort though is that you have become accustomed to how to address the bug reports when they are filed. Fortunately our V2 has also reached the point where we have been able to retire our bug reporting and pager setup as it now has its own implementation of this functionality. Although sometimes the reports are less than ideal, such as this evening what it filed a bug report, very loudly, 30 Min after going to bed that it was wearing the wrong pyjamas and the correct ones being it’s Toy Story pyjamas, which were in the wash…
- icambron 5y ago
  The maintenance of several products in different lifecycle phases is a huge issue for us. Different failure modes, different protocols for providing feedback, and different basic maintenance needs. For example, v1 requires the users to manage reading homework, the v2 requires bedtime stories, v3 requires diaper changes and milk, all simultaneously. This adds a lot of operational complexity for our already overwhelmed team, since they have to triage several different issue streams and context switch between them.
  - samwillis 5y ago
    We were concerned about trying to manage more than two of these products at once and so, as the Assistant to the VP of Reproduction, have made certain permanent operational changes to the equipment to ensure no further acquisitions are possible.
    - icambron 5y ago
      My cofounder and I were doing a recent retrospective on our most recent acquisition, and while we're very proud of the product, we considered that your approach may have been more sound.
      - samwillis 5y ago
        Well, if anyone is considering making similar changes to the equipment, in our experience it was very quick to implement with no side effects after about two days.
    - gumby 5y ago
      I literally do refer to that (male model) as the “hardware upgrade”
  - kimburgess 5y ago
    At least you’re working with versioned releases. I’ve recently acquired a new model and it’s definitely still in beta. There’s no documentation, conflicting user training and undefined behaviours everywhere. As soon as you discover workflows that meet your needs there appears to be a background update that while introducing new features, completely changes the behaviour of previous ones. Oh and don’t get me started on the sleep mode, it’s more unreliable that a Linux laptop and plays this horrible wake sound with no clear way to disable it. From what I’ve read on support forums this is happening to other users too.
    - icambron 5y ago
      They're really different product lines because each one is a complete rewrite; each has its own patch stream, which, like yours, change wildly without warning, as if the developers are just trying random stuff out. And nothing is ever backported between the products. While our different products have wildly different operational characteristics (and even move into different lifecycle phases at different times), the good news is that our team has gotten better at managing the constant flow of UX inconsistencies and changes. For example, my advice on sleep mode failures is to use the Erlang method: let it crash. We don't even look at the monitoring dashboards for our newest product anymore.
      - gumby 5y ago
        The problem is just when you get good at managing these things you realize production has stopped and you can’t put the skills to use any more.

samwillis 5y ago

The trouble we find is our v2 model, which we had hoped could have made improvements over v1, frequents the same unsavoury places that v1 did initially and continues to be compromised by the same vulnerabilities. It seems later versions are destined to make the same mistakes again. The only comfort though is that you have become accustomed to how to address the bug reports when they are filed. Fortunately our V2 has also reached the point where we have been able to retire our bug reporting and pager setup as it now has its own implementation of this functionality. Although sometimes the reports are less than ideal, such as this evening what it filed a bug report, very loudly, 30 Min after going to bed that it was wearing the wrong pyjamas and the correct ones being it’s Toy Story pyjamas, which were in the wash…

icambron 5y ago
The maintenance of several products in different lifecycle phases is a huge issue for us. Different failure modes, different protocols for providing feedback, and different basic maintenance needs. For example, v1 requires the users to manage reading homework, the v2 requires bedtime stories, v3 requires diaper changes and milk, all simultaneously. This adds a lot of operational complexity for our already overwhelmed team, since they have to triage several different issue streams and context switch between them.
- samwillis 5y ago
  We were concerned about trying to manage more than two of these products at once and so, as the Assistant to the VP of Reproduction, have made certain permanent operational changes to the equipment to ensure no further acquisitions are possible.
  - icambron 5y ago
    My cofounder and I were doing a recent retrospective on our most recent acquisition, and while we're very proud of the product, we considered that your approach may have been more sound.
    - samwillis 5y ago
      Well, if anyone is considering making similar changes to the equipment, in our experience it was very quick to implement with no side effects after about two days.
  - gumby 5y ago
    I literally do refer to that (male model) as the “hardware upgrade”
- kimburgess 5y ago
  At least you’re working with versioned releases. I’ve recently acquired a new model and it’s definitely still in beta. There’s no documentation, conflicting user training and undefined behaviours everywhere. As soon as you discover workflows that meet your needs there appears to be a background update that while introducing new features, completely changes the behaviour of previous ones. Oh and don’t get me started on the sleep mode, it’s more unreliable that a Linux laptop and plays this horrible wake sound with no clear way to disable it. From what I’ve read on support forums this is happening to other users too.
  - icambron 5y ago
    They're really different product lines because each one is a complete rewrite; each has its own patch stream, which, like yours, change wildly without warning, as if the developers are just trying random stuff out. And nothing is ever backported between the products. While our different products have wildly different operational characteristics (and even move into different lifecycle phases at different times), the good news is that our team has gotten better at managing the constant flow of UX inconsistencies and changes. For example, my advice on sleep mode failures is to use the Erlang method: let it crash. We don't even look at the monitoring dashboards for our newest product anymore.
    - gumby 5y ago
      The problem is just when you get good at managing these things you realize production has stopped and you can’t put the skills to use any more.