If you're not interested in games, why the interest in a Valve tablet?

SteamOS is probably the most commercially well-supported consumer desktop Linux, but it's still primarily a games storefront. If you only want to do desktop stuff, you're probably better off either on the Chromebook/Android train, or putting a community-supported distro on whatever hardware you've got.

You can put their OS on a minisforum V3 and that's like ... exactly what you want?

Part of the point and usefulness is having a stable target for developers to aim at, that they can test performance on. Also, most phones these days are roughly equivalent from the end-user perspective to ones from 2 or 3 years ago, the only difference is increased waste. So... no, no thank you.

Does anyone want to buy a phone every few years? No, I don't think they do.

It was released February 2022, that's only almost 4 years ago. 4-5 years is a good target for a refresh, I'll be somewhat surprised if there's not a new one in 2027 (but I was surprised by the lifespan of the Switch, and even the 7-8 years of the 360/PS3 era were surprisingly long, long generations are common now so no new Deck until 2028 or 2029 isn't out of the question), but any more frequently doesn't really make sense as the important components aren't improving in price/capability fast enough, and the initial release was and still is very capable rather than woefully inadequate. The motivations for upgrading are also different from a phone or more general laptop. I think the most common ranking of priorities for improvement would be: having various games run at all (mostly a software problem, Steam Deck already supports hardware ray tracing that various games now require), similar price range, better active battery life, physically lighter, and last would be higher graphical fidelity/performance. The things further down can't compromise the things higher up. Battery life advances being slow is kind of the killer.

There's a point that they could prioritize selling to new owners over existing owners looking to upgrade, and having a more capable device would help with that, but I think the marginal increase is probably not very big. The Steam Deck estimated sales were at 4 million units earlier this year, but that's still a relatively small portion of the whole PC gaming market (132m monthly active users on steam alone by 2021). It has been a big success for them, but it still exceeded their expectations, so I think they also would be skeptical of any large marginal improvement of new owner sales for what would likely be a minor improvement on the important specs. There's also competition from Windows handhelds whose sales don't suggest a large market just wishing Valve had a slightly more capable device that they'd pay more for.

A counter argument - the Switch gave game devs a solid platform to target without being the latest and greatest without compromising the usability or fun factor

I would much rather a refresh every 5+ years with a more profound hardware improvement. I'm even fine with closer to 10 years if the technology hasn't changed that much at the 5 year mark.

Why should they? Do you think the phone refresh cycle is a healthy one to emulate?

No one would bother optimizing for it at all if it had a yearly refresh.

Not saying that's why they don't, but that would be a side effect. I actually think it's more their business doesn't rely on selling Steam Decks, hardware business where you do a yearly refresh is a very different beast to one you do a new model every few years. Their organization doesn't seem set up for that.

I watched some behind the scenes videos about Valve’s Steam Frame development, and it doesn’t seem like they have a very big hardware team.

It's been just 2 years since the OLED release, I think we're closing in on a refresh. Unless a deck is a year away from a generational bump. A refresh could include the updated joysticks featured on the Steam controller, though.

Till then I'd think I'd do more good for Valve to focus on their steam app and store experience.

honestly at this point, phones and personal computers probably should move to a 2 year cadence. The R&D costs are going up and the performance benefits are decreasing.