Damien MargaritisI don't work for Microsoft so their motives are irrelevant to me personally, but in general companies have a fiduciary responsibility to up-sell and one of the ways they up-sell is by obfuscating their reasoning with unnecessary technical pleas. For example, anywhere you see justifications about approaches taken, they're not talking about literal technical limitations like a "limited by the technology of my time" howard stark meme. They're talking about selling you on the most expedient way they could accomplish their specific goals while costing them the least amount of money to generate the solution. in the case of your blog, let's just hit the bullet points to find all the sales pitches that are happening.
"control end user experience" different is just different, not better or worse. Microsoft was actually reverse selling this feature not more than 2-3 years ago by claiming that each vendor was able to apply more aggregate resources and work in a competitive landscape to pull off 3pip interfaces that MS never would have been able to achieve by continuing the LPE program in-house. So you can see this bullet point is 100% arbitrary depending on whatever story Microsoft is telling at the time.
"lack of feature parity" again this was being heralded as a positive in that it provided customers who needed more features to choose more expensive vendor solutions and customers who needed cheap phones the ability to give up some features. so basically how great flexibility and choice was in the crucible of the free market.
"Inconsistent delivery of new features" you had to push out new firmware whether it was tanjay/aries or 3pip so obviously you're not talking about that, but the functionality being pushed around in o365 tenants is anything but consistent so I know you're not taking that angle. as for your naked assertion that it's maybe sorta not technically possible to code new features into existing firmware that has already been established as FUD or at bare minimum an additional development branch that Microsoft simply doesn't want to manage as "it doesn't fit with their approach" aka it will either be too much hassle or cost too much.
"there's no SIP in teams" so what? you wrote two paragraphs that boil down to "you have to use a different protocol to register." so is the subtext here that developers are incapable of adding teams registration to a 3pip firmware? because I'd absolutely love to hear a polycom or yealink dev admit that in a public forum. Or are you saying the sip gateway to cloud voice is already in maintenance mode and microsoft doesn't want to develop against it anymore? because that goes right back to hassle/money again.
Now just to bring it all home, I'm 100% sure that Microsoft thinks they are making the best decisions for managing Teams as a live service over the next 10 years, but they've also over-promised and under-delivered in regard to voice services and supportability matrices in that area, and not just with their constantly changing definitions of "certified/optimized for Skype fully supported I mean mostly supported ok fine here's a subset of features that will work" and "LPE supported for Teams woops not supported afterall" and "TLS 1.0 actively blocked woopsie we just meant not supported" and their own personal mission accomplished variation "now has feature parity with SFBO".
In a vacuum while reading about it in news articles and blogs it's all well and good but when you've got your MS rep squeezing you on how much better Teams phones are less than 2 years after replacing all LPE phones with 3PIP phones and then you have fasttrack consultants trying to change your plans just because Microsoft added a Skype CU one month ago that provided a new migration path and then you can't even get your PoC on how to handle 3pip phones interop with call queues definitively answered because now Microsoft doesn't want you to migrate SFB Server to SFBO before upgrading to teams because they want to deprecate SFBO... ok well that's all really annoying, but they haven't provided any guidance on migrating RGS agents to Teams call queues much less whether or not those agents will need new hardware or not.
So then you come to the realization that this is all done because they just really don't care that they were selling 2 year lifecycle phones and/or there is some program manager somewhere who has said in a private meeting, "they'll just have to suck it up and buy new phones if they want xyz" while completely ignoring the number of hoops that customers had to jump through just to get their 3pip phones purchased. And then you come across some post that talks about how this situation "isn't ideal" and that this consolidation in direct opposition to what they were selling not very long ago "will open up many more capabilities" while failing to name a single capability other than that it will make the Microsoft devs' lives easier. Then yes, at that point I will go out of my way to say that at a very fundamental level these are all arbitrary excuses that sit on a throne of lies. and at that same level, the reason you're getting limited backwards compatibility is due to both money and hassle or some variation thereof so as far as I'm concerned coming from the perspective of a Microsoft customer I'm completely justified in my irritation regardless of their reasoning for doing what they've done. It's the ends that are the problem here, not the means.