4 ms·

> perceived stability This is begging the question, though. I'll bet that the average company can't maintain the same uptime as a cloud solution, even with hig

by Cpoll 5y ago

> perceived stability This is begging the question, though. I'll bet that the average company can't maintain the same uptime as a cloud solution, even with high-profile one-week outages like these. > as soon as you switch platform Implying that companies don't switch platforms when they have on-prem installations. > I have systems in production that nobody touched for half decade But that's apples and oranges. The comparison isn't Atlassian Cloud vs. your undefined production system, it's Atlassian Cloud vs. Atlassian On-prem (or $PLANNING_SOFTWARE_OF_CHOICE).

majkinetor 5y ago
> I'll bet that the average company can't maintain the same uptime as a cloud solution, even with high-profile one-week outages like these. Its not only about uptime, its also about maintenance that is required to keep that uptime. It isn't there forever, you need to follow and adapt to API, license and other changes continuously. > Implying that companies don't switch platforms when they have on-prem installations. Be that as it may, you typically choose something that is not invented few years ago, but typically exists for decade and is battle tested. Lots of generic knowledge too, usable in other environments. And its not like people don't change Cloud platforms either... > The comparison isn't Atlassian Cloud vs. your undefined production system, it's Atlassian Cloud vs. Atlassian On-prem (or $PLANNING_SOFTWARE_OF_CHOICE). Comparison isn't between different supportive systems either. If I have a service, which is supported by specific infrastructure, I care about a service and current infra is just one mean to that end (I will happily replace parts of it that aren't compatible with that goal). The comparison is a about which supportive system makes me do less work on the long run while keeping the same level of SLA. I don't care about individual features of supportive system in that regard, I care just about end effect. For one, if I have to relicense stuff every few months, its immediately not that good, as I can choose FOSS and let it be for decade.
mbrodersen 5y ago
If there is a problem with infrastructure at my company, I can go and talk to one of the sys admins and get it sorted right away. I have never experienced outages as bad as what we are often seeing with cloud providers.

by Cpoll 5y ago

majkinetor 5y ago

> I'll bet that the average company can't maintain the same uptime as a cloud solution, even with high-profile one-week outages like these. Its not only about uptime, its also about maintenance that is required to keep that uptime. It isn't there forever, you need to follow and adapt to API, license and other changes continuously. > Implying that companies don't switch platforms when they have on-prem installations. Be that as it may, you typically choose something that is not invented few years ago, but typically exists for decade and is battle tested. Lots of generic knowledge too, usable in other environments. And its not like people don't change Cloud platforms either... > The comparison isn't Atlassian Cloud vs. your undefined production system, it's Atlassian Cloud vs. Atlassian On-prem (or $PLANNING_SOFTWARE_OF_CHOICE). Comparison isn't between different supportive systems either. If I have a service, which is supported by specific infrastructure, I care about a service and current infra is just one mean to that end (I will happily replace parts of it that aren't compatible with that goal). The comparison is a about which supportive system makes me do less work on the long run while keeping the same level of SLA. I don't care about individual features of supportive system in that regard, I care just about end effect. For one, if I have to relicense stuff every few months, its immediately not that good, as I can choose FOSS and let it be for decade.

mbrodersen 5y ago

If there is a problem with infrastructure at my company, I can go and talk to one of the sys admins and get it sorted right away. I have never experienced outages as bad as what we are often seeing with cloud providers.