3 ms·

> Users would rather have [bloated JS webmail] vs a plain HTML page with links. The empirical evidence is clear on that, otherwise the web would not have evolve

> Users would rather have [bloated JS webmail] vs a plain HTML page with links. The empirical evidence is clear on that, otherwise the web would not have evolved as it did. There's got to be a special name for this kind of logical fallacy. > Software is built on layers of abstraction, and necessarily, there will be bloat as a byproduct of layering. Except we're layering less powerful, less abstracted APIs over an already high-level document model, or we're layering a primitive record access API over SQL.

anewhnaccount2 8y ago
>There's got to be a special name for this kind of logical fallacy. Similar to https://en.m.wikipedia.org/wiki/Just-world_hypothesis https://en.m.wikipedia.org/wiki/Just-world_hypothesis perhaps?
shaki-dora 8y ago
> There's got to be a special name for this kind of logical fallacy. There also needs to be a Fallacy fallacy: the irrational believe that merely dropping the "fallacy" moniker is an argument. In this case: Yes, considering there are many hosted email services competing for customers, with quite a lot of variation between "huge web app" and "minimalistic list of links". And considering Google is known to excessively A/B test any change (including more-complex v. just-html interfaces), it's quite valid to conclude that people generally prefer what GMail is doing. GMail even has a fallback html interface it switches to when noticing slowdowns. The number of people keeping that mode on manually (few) is another indicator of people's preferences. > Except we're layering less powerful, less abstracted APIs over an already high-level document model[...] "Abstract" != "good". Abstraction layers can be stupid, or deliberately constricted. They are still, tautologically, by definition, one step more "abstract" than the API they access.
TeMPOraL 8y ago
> There's got to be a special name for this kind of logical fallacy. Voting-with-your-wallet fallacy? Users choose from what's available on the market. Between high marketing, network effects and lack of technical understanding of the average user, there is close to zero feedback going back to service providers. The providers get to unilaterally decide what's on the market, and users have no choice but to take it.

anewhnaccount2 8y ago

>There's got to be a special name for this kind of logical fallacy. Similar to https://en.m.wikipedia.org/wiki/Just-world_hypothesis https://en.m.wikipedia.org/wiki/Just-world_hypothesis perhaps?

shaki-dora 8y ago

> There's got to be a special name for this kind of logical fallacy. There also needs to be a Fallacy fallacy: the irrational believe that merely dropping the "fallacy" moniker is an argument. In this case: Yes, considering there are many hosted email services competing for customers, with quite a lot of variation between "huge web app" and "minimalistic list of links". And considering Google is known to excessively A/B test any change (including more-complex v. just-html interfaces), it's quite valid to conclude that people generally prefer what GMail is doing. GMail even has a fallback html interface it switches to when noticing slowdowns. The number of people keeping that mode on manually (few) is another indicator of people's preferences. > Except we're layering less powerful, less abstracted APIs over an already high-level document model[...] "Abstract" != "good". Abstraction layers can be stupid, or deliberately constricted. They are still, tautologically, by definition, one step more "abstract" than the API they access.

TeMPOraL 8y ago

> There's got to be a special name for this kind of logical fallacy. Voting-with-your-wallet fallacy? Users choose from what's available on the market. Between high marketing, network effects and lack of technical understanding of the average user, there is close to zero feedback going back to service providers. The providers get to unilaterally decide what's on the market, and users have no choice but to take it.