4 ms·

I have a MacBook air 13", the basic one with 4GB of ram and a small SSD. It works fine. I can't really see why it would need to get a whole lot better honestl

I have a MacBook air 13", the basic one with 4GB of ram and a small SSD. It works fine. I can't really see why it would need to get a whole lot better honestly. If I need a "big" computer then I use my desktop, and if I just want to read then I use my tablet or my phone. I would buy another apple laptop sure but not because of the specs really. They are just good computers in a world where there are lots and lots of good computers. I think apple is right to just keep on keeping on with these machines anyway because there isn't much innovation in terms of parts happening now. Touchscreens are a gimmick to me and newer cpus are only a little better and not worth the price of the upgrade. New computer specs are basically about as exciting as reading about innovations in Refrigerators or other durable goods.

swozey 10y ago
The goal for a lot of us is to not need a "desktop" to do our "big" work. That's the thing. This was the Macbook Pro line, meaning you'd carry it home and to work and plug it into a dock/monitors and never need a "big" computer.
- tgarma1234 10y ago
  In that case it seems like using "big" desktop computers on Azure or AWS is right around the corner, especially if you factor in better internet speeds to make the remote desktop experience more seemless, so that the client computer doesn't need all that much power. It just doesn't seem like there is a future in developing computers with massive specs as personal machines anymore. If there was a market in it that was profitable, then the analysts at all of the major companies would know it. They seem to see that there isn't a market worth investing in.
  - swozey 10y ago
    There is a HUGE monetary and time benefit to having developers spin up docker/vagrant environments to run integration/etc tests on their laptops before it hits CI/CD systems so I completely disagree. People thought we'd offload developer computing to the "cloud" 5 years ago. It's not coming soon, I doubt it's coming in the next 5-10 years if you're making applications that now rely on microservices, kubernetes, etc. AWS charges by the hour, GCP charges in 1 minute increments after the first 10 minutes. That all is extremely costly if you have a number of developers sending all of their tests through those systems.
    - tgarma1234 10y ago
      And that seems to me like the main incentive they would have to force developers into that user model. On the one hand it makes sense to charge $2500 for a laptop but then if the quality is so great now that the machine can be expected to work well for almost every user for nearly 10 years then that price point is actually almost a Pyrrhic victory. It would be better to sell smaller cheaper machines that justify upgrades more frequently and then charge power users for time in the cloud. I think that's pretty much the inevitable endgame here in an industry that is all but fully commoditized.

swozey 10y ago

The goal for a lot of us is to not need a "desktop" to do our "big" work. That's the thing. This was the Macbook Pro line, meaning you'd carry it home and to work and plug it into a dock/monitors and never need a "big" computer.

tgarma1234 10y ago
In that case it seems like using "big" desktop computers on Azure or AWS is right around the corner, especially if you factor in better internet speeds to make the remote desktop experience more seemless, so that the client computer doesn't need all that much power. It just doesn't seem like there is a future in developing computers with massive specs as personal machines anymore. If there was a market in it that was profitable, then the analysts at all of the major companies would know it. They seem to see that there isn't a market worth investing in.
- swozey 10y ago
  There is a HUGE monetary and time benefit to having developers spin up docker/vagrant environments to run integration/etc tests on their laptops before it hits CI/CD systems so I completely disagree. People thought we'd offload developer computing to the "cloud" 5 years ago. It's not coming soon, I doubt it's coming in the next 5-10 years if you're making applications that now rely on microservices, kubernetes, etc. AWS charges by the hour, GCP charges in 1 minute increments after the first 10 minutes. That all is extremely costly if you have a number of developers sending all of their tests through those systems.
  - tgarma1234 10y ago
    And that seems to me like the main incentive they would have to force developers into that user model. On the one hand it makes sense to charge $2500 for a laptop but then if the quality is so great now that the machine can be expected to work well for almost every user for nearly 10 years then that price point is actually almost a Pyrrhic victory. It would be better to sell smaller cheaper machines that justify upgrades more frequently and then charge power users for time in the cloud. I think that's pretty much the inevitable endgame here in an industry that is all but fully commoditized.