5 ms·

Ask HN: Will you use software that has telemetry on by default?

10 points by bluefishinit 3y ago

ActorNightly 3y ago
Don't care as long as it doesn't affect core functionality (like requires internet connection to use for something that should work offline). The benefit of a product that is paid for by personal data is immediate and real to me - I use this product and get benefit about it. The risk of said personal data being used against me is currently hypothetical. So far, I haven't experienced any drawbacks. For anything that is privacy sensitive, I use a dedicated privacy oriented software.
JohnFen 3y ago
If the software isn't critical, then no. Making telemetry opt-out is making a statement about the attitude of the producers that makes me very uncomfortable. If the software is critical, then yes. And I'll opt out. And I'll keep scouting around for a suitable replacement.
satvikpendem 3y ago
Yes, because I don't care about telemetry. I have telemetry in the products our company makes because it's a useful tool to see where customers are stuck. At the end of the day, those customers pay for our salaries so we'll do whatever we can to make them happy.
- bluefishinit 3y ago
  > customers pay for our salaries so we'll do whatever we can to make them happy To a lot of people that means no telemetry. Look at what happened with the Go compiler telemetry. I think that's the only time I've seen someone actually listen to customer desire and remove telemetry.
  - satvikpendem 3y ago
    It's a good thing I don't make developer products then. It's wild how devs will literally use telemetry data to fix bugs in their employer's products yet won't give the same affordances to devs of other companies.
    - bluefishinit 3y ago
      > It's wild how devs will literally use telemetry data to fix bugs in their employer's products Fwiw I'd never work somewhere that was doing telemetry driven development, so at least in my case this isn't true.
moomoo11 3y ago
As long as it’s used to improve the product what’s wrong with that? Are product people supposed to chant and perform esoteric rituals to gain telekinetic insight into how the product is being used?
- bluefishinit 3y ago
  > As long as it’s used to improve the product what’s wrong with that? I see it as a violation of privacy. I also don't think it leads to good products. > Are product people supposed to chant and perform esoteric rituals to gain telekinetic insight into how the product is being used? They should be knowledgeable in their field and leverage their experience to solve problems with the product. No telemetry was used to create the iPhone for instance, it was a product built with a strong vision. When you do telemetry driven development it's super easy to get trapped in local maxima. It also removes the creative process from product design.
  - moomoo11 3y ago
    I guess I’m more thinking about enterprise software products than physical products.
    - bluefishinit 3y ago
      I would say that consumer products and physical products are similar in that the bar is very high for UX and innovation so an expert with a point of view is the best option. Enterprise product features are best off led by sales. What helps sell? What are customers complaining about? The UX bar is low for enterprise so the "too many cooks in the kitchen" model is passable there as long as the functionality matches the market.
zzo38computer 3y ago
Preferably, only if it is FOSS and I can disable the telemetry before starting the program for the first time. Especially for local software. If the software requires the internet anyways, it is only acceptable if it is only for internet related functions and only if it connects to the servers that I will specify (e.g. if it is a NNTP client, to be able to specify my own NNTP servers including LAN or localhost if I want to, even when it is not actually connected to the internet). For knowing how to improve the software, I would like to write public comments to criticize it so that they can improve it from that (and so that others can publicly criticize my public criticism too if desired).
bluefishinit 3y ago
The answer for me is a hard no, but I'm curious what others think. The justification often used is that it's the only way to "make the product better". That's not been my experience as either a developer or user. My favorite tools are opinionated and created by people with a desire to solve a problem and good UX skills (think git, vim...). My least favorite tools are corporate software that was obviously built and designed by committee.
- LinuxBender 3y ago
  This could be a good use of HN's polling system [1]. To answer your question I would not and if required to use it I would make a serious attempt to block or spoof it make it think it's sending data to the right place. [1] - https://news.ycombinator.com/newpoll https://news.ycombinator.com/newpoll
- rolph 3y ago
  if i was bored, i would lift the telemetry monster out, and subvert it to send spoofs.
- Maro 3y ago
  You also don't use anything in the Google ecosystem, or Facebook, or any online apps? Those all have telemetry..
  - bluefishinit 3y ago
    I use them but I would count them as some of the worst products I'm forced to interact with.
Corsome 3y ago
Only as a last resort when no alternative is available. Fortunately this is rare since we've got plenty of options available.
countWSS 3y ago
I switched to Librewolf just because of Firefox telemetry. So the answer is generally no, i will however try your software in networkless VM as last resort if alternatives don't exist.
eternityforest 3y ago
Telemetry doesn't factor into my choice of software much. I would guess at least half my computer time is spent in telemetry heavy apps.
khedoros1 3y ago
I try not to; there's usually a good alternative.
add-sub-mul-div 3y ago
It's possible I've let myself develop too negative a knee jerk reaction to the idea of telemetry. Sometimes I've lost a feature I use in a product that I'd turned off the telemetry for, and the developers cited low usage of that feature. Did I shoot myself in the foot?
- JohnFen 3y ago
  I figure that if the devs are making decisions about what features to keep or drop based mostly on telemetry signals, that alone is enough reason to avoid using the software.
korpsey 3y ago
Depends if is it on by default but the project is open about it or if it's "hidden". I usually try to avoid the later.
marssaxman 3y ago
Not if I have any choice in the matter - and more often than not, I do.

10 points by bluefishinit 3y ago

Don't care as long as it doesn't affect core functionality (like requires internet connection to use for something that should work offline). The benefit of a product that is paid for by personal data is immediate and real to me - I use this product and get benefit about it. The risk of said personal data being used against me is currently hypothetical. So far, I haven't experienced any drawbacks. For anything that is privacy sensitive, I use a dedicated privacy oriented software.

JohnFen 3y ago

If the software isn't critical, then no. Making telemetry opt-out is making a statement about the attitude of the producers that makes me very uncomfortable. If the software is critical, then yes. And I'll opt out. And I'll keep scouting around for a suitable replacement.

satvikpendem 3y ago

Yes, because I don't care about telemetry. I have telemetry in the products our company makes because it's a useful tool to see where customers are stuck. At the end of the day, those customers pay for our salaries so we'll do whatever we can to make them happy.

bluefishinit 3y ago
> customers pay for our salaries so we'll do whatever we can to make them happy To a lot of people that means no telemetry. Look at what happened with the Go compiler telemetry. I think that's the only time I've seen someone actually listen to customer desire and remove telemetry.
- satvikpendem 3y ago
  It's a good thing I don't make developer products then. It's wild how devs will literally use telemetry data to fix bugs in their employer's products yet won't give the same affordances to devs of other companies.
  - bluefishinit 3y ago
    > It's wild how devs will literally use telemetry data to fix bugs in their employer's products Fwiw I'd never work somewhere that was doing telemetry driven development, so at least in my case this isn't true.

moomoo11 3y ago

As long as it’s used to improve the product what’s wrong with that? Are product people supposed to chant and perform esoteric rituals to gain telekinetic insight into how the product is being used?

bluefishinit 3y ago
> As long as it’s used to improve the product what’s wrong with that? I see it as a violation of privacy. I also don't think it leads to good products. > Are product people supposed to chant and perform esoteric rituals to gain telekinetic insight into how the product is being used? They should be knowledgeable in their field and leverage their experience to solve problems with the product. No telemetry was used to create the iPhone for instance, it was a product built with a strong vision. When you do telemetry driven development it's super easy to get trapped in local maxima. It also removes the creative process from product design.
- moomoo11 3y ago
  I guess I’m more thinking about enterprise software products than physical products.
  - bluefishinit 3y ago
    I would say that consumer products and physical products are similar in that the bar is very high for UX and innovation so an expert with a point of view is the best option. Enterprise product features are best off led by sales. What helps sell? What are customers complaining about? The UX bar is low for enterprise so the "too many cooks in the kitchen" model is passable there as long as the functionality matches the market.

zzo38computer 3y ago

Preferably, only if it is FOSS and I can disable the telemetry before starting the program for the first time. Especially for local software. If the software requires the internet anyways, it is only acceptable if it is only for internet related functions and only if it connects to the servers that I will specify (e.g. if it is a NNTP client, to be able to specify my own NNTP servers including LAN or localhost if I want to, even when it is not actually connected to the internet). For knowing how to improve the software, I would like to write public comments to criticize it so that they can improve it from that (and so that others can publicly criticize my public criticism too if desired).

bluefishinit 3y ago

The answer for me is a hard no, but I'm curious what others think. The justification often used is that it's the only way to "make the product better". That's not been my experience as either a developer or user. My favorite tools are opinionated and created by people with a desire to solve a problem and good UX skills (think git, vim...). My least favorite tools are corporate software that was obviously built and designed by committee.

LinuxBender 3y ago
This could be a good use of HN's polling system [1]. To answer your question I would not and if required to use it I would make a serious attempt to block or spoof it make it think it's sending data to the right place. [1] - https://news.ycombinator.com/newpoll https://news.ycombinator.com/newpoll
rolph 3y ago
if i was bored, i would lift the telemetry monster out, and subvert it to send spoofs.
Maro 3y ago
You also don't use anything in the Google ecosystem, or Facebook, or any online apps? Those all have telemetry..
- bluefishinit 3y ago
  I use them but I would count them as some of the worst products I'm forced to interact with.

Corsome 3y ago

Only as a last resort when no alternative is available. Fortunately this is rare since we've got plenty of options available.

countWSS 3y ago

I switched to Librewolf just because of Firefox telemetry. So the answer is generally no, i will however try your software in networkless VM as last resort if alternatives don't exist.

eternityforest 3y ago

Telemetry doesn't factor into my choice of software much. I would guess at least half my computer time is spent in telemetry heavy apps.

khedoros1 3y ago

I try not to; there's usually a good alternative.

add-sub-mul-div 3y ago

It's possible I've let myself develop too negative a knee jerk reaction to the idea of telemetry. Sometimes I've lost a feature I use in a product that I'd turned off the telemetry for, and the developers cited low usage of that feature. Did I shoot myself in the foot?

JohnFen 3y ago
I figure that if the devs are making decisions about what features to keep or drop based mostly on telemetry signals, that alone is enough reason to avoid using the software.

korpsey 3y ago

Depends if is it on by default but the project is open about it or if it's "hidden". I usually try to avoid the later.

marssaxman 3y ago

Not if I have any choice in the matter - and more often than not, I do.