When the heck will Home Assistant 1.0 / Home Assistant Core 1.0.0 be released?
Some might say that is only numbers on paper but I believe that releasing 1.0 of Home Assistant / Home Assistant Core would mean so much more in the hearts and souls of HA users.
It could have a significant impact on how Home Assistant is seen by the media and new users that might be hesitant to try out a software application which does not yet even have a “1.0” release. More public reviews should lead to more users and that could lead to more developers which in turn would probably mean a better HA for all of us.
Personally I think it is important just in order to more clearly see that the project is moving forward, it simply makes to feel proud to publicly see the project you use and cheer for making progress when you read about them on the larger tech-news sites.
Version 1.0 of a software application was historically used as a major milestone, indicating that the software is “complete”, however, today we know that no software is really ever “complete”.
These days many projects that follow a rolling releases schedule use announcement of major release versions as more in the spirit of celebrating the feeling of accomplishment, such example that “a lot” of new features and/or functions have been merged within a time period agreed upon that still remains flexible.
You could, for example, agree to try to announce a major version release approximate every 18-24 months no matter what features has been merged.
Ironically large commercial software companies, like Google, Microsoft and others, release unfinished products all the time just to get first-to-market, even Enterprise class products with very expensive software license fees.
Not to mention new large games from big video game companies requiring day-one-patches
It’s a risk but I guess their PR/Marketing-departments still follow the maxim that any publicity is good publicity, though research point to that only applying to small or unkown companies with nothing to loose
“Most companies are concerned with one of two problems,” says Alan Sorensen, associate professor of economics and one of the study’s authors. “Either they’re trying to figure out how to get the public to think their product is a good one, or they’re just trying to get people to know about their product. In some markets, where there are lots of competing products, they’re more preoccupied with the latter. In that case, any publicity, positive or negative, turns out to be valuable.”
Which might never happen, because goals defined today might lost importance in next few years. Or there will be technology changes which makes HA implicitely outdated (ie not deserving 1.0 version number).
We will get there eventually. While goals will always be added and more will be wanted. The main goal is to get to a place where we create the best experience for a user with the most features and usability of the UI. From a frontend perspective, I think we have a bit more to do to make the UI more intuitive and so the user gets the most from the experience.
It took Inkscape 17 years to get to an initial 1.0 release. It took GIMP 3 years. Both aforementioned software’s had breaking changes after 1.0. So what’s my point? Why rush 1.0 when it really doesn’t matter because there will still be breaking changes.
If it doesn’t matter, why to not call it v1? Don’t we actually call it v114?
BTW current version (lower than 1) is often exploited in discussions being thrown as argument. Which is pretty weak one, since there will be the same issues and discussions after 1.0.
What’s the point in making the change if there’s already a plan? I wouldn’t want to invest more time into a new plan just because some people wanted less letters characters.
That’s usually to stifle complaints about breaking changes. We’ve all dealt with them, no reason to have incoherent rants about why they shouldn’t exist.
What plan? 5-years one? This one with all those vague goals? Do we do waterfall over here?
I cannot see the plan. There are some general milestones or ideas which could change over time anyway… Nobody knows what will happen in year or two and what requirements will be on the table at this time.
But you cannot guarantee that there will be no breaking changes after releasing v1.0. In this context such argumentation is an excuse only. In other words v1.0 will not be different in regard of development pace nor breaking changes.
All this debate circles back to it not making a difference what version we are on. So what’s the point in doing extra work if they are effectively the same? None. To make (currently) 6 people happy just because is pointless.
IMO reasons “why” have been said already. Number of votes used by you as argument in discussion is a proof you didn’t even try to understand the motivation.
So maybe it would be better you will not try to convince us. Just leave it and go over. I’m sure project owners can decide on their own their further steps either based on inputs found in this thread or not.
1.0 gives a sense that Home Assistant is “done” to me. Though it will never be done. I don’t think we are to a point of saying we are ready to say HA is 1.0 ready. This is not really something that should be decided by the community IMO. (this is not necessarily the opinion of HA or Nabu Casa employees as I am not one of them.) This should be from the HA members and the Maintainers (employees of NC) as it is their “product”. They should say when they think it is good enough. Again just my opinion
I understand the reasoning and I don’t agree with it. I even used 2 different open source products as an example, both which are widely used and both took different paths. I stated that it didn’t make a difference for each package. Essentially refuting OPs reasoning. Maybe I needed to be more explicit with you for you to understand that?
Anyway, there will always be breaking changes. As manufacturers find new ways to break standards (whether actual or de facto) then HA devs will have to adapt.Sometimes adapting means breaking with past code, or refactoring it.
I mean, once there was essentially just X10 in this space, and some horribly expensive proprietary s**t. If someone had designed HA around X10, and released it as 1.0, then there would have been breaking changes for Z-Wave, zigbee and all the other product integrations. This is an extreme example, but OTOH once upon a time no one had dreamed of the easy and cheap availability of most IoT technologies. And no one knows what will come in 6 months or 2 years.
I am all in favour of getting the basics right - the devs from balloob down are learning the best way to do things, and learning from past coding practices.
It’s just a number but it signifies it’s ‘done’ and ready for moms and pops and it’s not there yet. I don’t see why there should be any rush to get some feel good v1 just because. Actually I remember a year ago when 0.99.x was out the devs DID want to release 0.100 as 1.0 but it wasn’t ready then and still isn’t. And so what? This has been an awesome year of great releases and 0.115 looks like being a blockbuster release…
