For production workloads, you can use systemd socket activation to avoid most of the network issues. The caddy demo I've linked below explains more about the issues it would solve.
I remember the preview price for 2.5 flash was much cheaper. And then it got quite expensive when it went out of preview. I hope the same won't happen.
For 2.5 Flash Preview the price was specifically much cheaper for the no-reasoning mode, in this case the model reasons by default so I don't think they'll increase the price even further.
Will you get upset if Microsoft will charge 500000000 USD (because more copilot value added every month) per year? That is way more upsetting imho. And if all fails there is still some SAP solution to everything in life :P
This one gets me now than most of the rest... The increase in licensing for copilot features a lot of orgs would prefer to disable is distasteful to say the least.
I wonder what they use for Microsoft Office. My office license is renewing in 2 weeks and I have been looking at alternatives but they all have their own catch.
The catch depends a lot on the context that you're considering. Trying to replace Microsoft Office as a whole by a drop-in replacement like LibreOffie may work better or worse depending on who uses it.
I've never used anything but OpenOffice / LibreOffice for writing academic texts in the humanities and never missed anything. The "catch" whenever I tried Microsoft Word was the menu that had the most important functions (for me) hidden away much deeper than in OO and LO.
I've never been a big user of Spreadsheets but I've heard only good of Excel and trust the widespread opinion that it is unchallenged in its domain. In sociology you wouldn't use it because you've got specialized statistics software such as R and SPSS (PSPP being an attempt at an Open Source Alternative to SPSS).
Looking at administration, Excel ist probably quite important but when you get rid of it, not one but various solutions might take its place, depending on who uses it. If you want something like a browseable database in a colorful table for office clerks, LO Calc might be enough. But the things Excel gets praised for a lot (I never know what exactly people mean) would probably have to be tackled another way.
Governments going down that need to invest into finding those solutions by providing staff that is qualified to find them or even develop them. The state of Schleswig-Holstein considered in its Open Source initiative strategy that it may be challenged by a future legislation and put a focus on the reasons for acceptance of Open Source solutions. I wonder if that is put into action well to find solutions with the least "catch" that may even excel over Microsoft products depending on their context :)
Immerse yourself in a workplace setting where Excel is the first thing that people grab for anything but text editing. You'll see how insanely productive people are. Now actually try switching to LO Calc.
I've done this several times during my career, to see if LO Calc would ever come up to the performance of Excel. To be fair, I haven't done so since I switched to Python.
Here's the experiment I would conduct. Generate a column of 5000 numbers. Now graph them. Now make a few token changes to the graph such as modifying some of the aesthetic parameters. The difference in processing time was profound, last time I tried it. Also, there was a noticeable "latency" between clicking something, and seeing something happen, that made it quite un-ergonomic if not physically painful to use. I'm sensitive to this because I get eyestrain headaches easily.
I find Onlyoffice to be the closest alternative. It presents itself as a hosted office platform but you can actually install it locally and it feels just like an office program.
It's not the most efficient, being effectively a webview. But its UI and compatibility is imho much better than LibreOffice.
Super basic use cases, like having a CV in Word and a Budget Excel. But I switched to Onlyoffice now. I am not a huge fan of scripting in excel anyways.
It's not about outages. It's about the why. Hardware can fail. Bugs can happen. But to continue a roll out despite warning sings and without understanding the cause and impact is on another level. Especially if it is related to the same problem as last time.
You don't need outages to build experience in resolving them, if you identify conditions that increase the risk of outages. Airlines can develop a lot of experience resolving issues that would lead to plane crashes, without actually crashing any planes.
reply