No More Auto-Retiring Beers

I was actually a fan of auto-retiring, but then I came across a small local brewery that wasn't getting and probably wouldn't get the rating volume of other breweries. I had forgotten if a new rating would auto-unretire a beer (I don't think it does), but maybe it should (even if an occasional aged beer becomes "unretired" for a year).

It's likely too much work and a guessing game as to an actual algorithm, but I wonder if a brewery's overall volume of beer ratings (across all beers) might weight the length of time before a beer is considered "retired."

Anyhow, the just undone technique had its flaws, but I also think relying on the community is essentially doing away with retired beers. The last active data is nice to go one, and maybe that could be come more prominent and/or be adjusted/added to to include a brewery-specific version of "trending beers" filter/sort to help people find records for breweries with 100s of beers.

Mainly just spitballing some thoughts trying (and likely failing) to be constructive without complaints. Thanks for continuing to try out things to improve the user experience here.

 

Thank you! Real good update.
Will be great to add auto-retiring for beer with Zero Rating which was added two (for example) years ago.

 

Happy to be reading this- Thank you, @Todd!

 

Nice update, now we are nearer the truth for at least the beers I'm drinking.

 

I can’t say I was a fan of the last shake ups with “inactive” and whatnot. I thought the auto moving of beers to “retired” would be confusing for the user interface. Why would someone look in the “retired” section for a beer that was in production? I’ve seen posts on BA that expressed this confusion.

Active and retired are not opposites. Active refers to a beer’s status on BA, while retired should refer to a beer’s status with a brewer. In my opinion, the muddying of this is the issue.

The BA database is at its best when consumers are entering in info directly from a brewer - name, style, ABV… and adding their opinion about a beer as a consumer (the review). It’s less effective when making presumptions about a beer.

I felt like “retired” was always an issue. How do we know what’s retired? Beers leave a production schedule and then they come back. It feels like it’s frequently inaccurate. The auto-retiring created inaccuracies, but a manual retirement that relies on user/mod action would do the same.

Why not just change the name of “retired” to “inactive” and increase the amount of time it takes for a beer to automatically become “inactive” - like 3 years? This way BA is using BA activity to make appropriate labels for beers rather than trying to imperfectly make claims about a beer’s status with a brewer without the appropriate info.

@beer_beer ’s post above said something very telling: “now we are nearer the truth.” How about BA indicates BA’s truth rather than something only near a brewer’s truth?

Frankly, I don’t even see a significant issue if you take this to it’s logical extreme and have “inactive” beers automatically become “active” with a review (as long as there isn’t a “retired” category). If someone is rating an out of production beer that’s been aged at least 3 years, is it a bad thing that the beer appears in the active list? The active list should reflect user activity. If that’s problematic, then make the automation only kick in when an inactive beer gets 2 ratings within a year (or something to that effect).

In short - make a beer’s status reflect BA activity, have everything automated, and use terms that match this rather than ones that create a false impression.

I understand why some really wouldn’t like this change, but it just seems like BA’s use of “retired” is forcing a square peg into a round hole. The relevance of the term fit an earlier era in beer and isn’t as applicable today. Sometimes the best solution is the most straightforward and clean one.

 

I like the term "Out of Production"; it covers lots of ground. As a specific example, look at a brewery like Spencer. Closed but still has 18 'active' beers on this site. And I know for a fact that there is still some of their beer in the distribution chains. So closed does not mean retired or inactive but it does mean they are not being produced.

 

I don't know how things work behind the scenes, so if I go thru the beer list of a local brewery and create a list of actual retired beers, is it necessary to update these beers one at a time on that beer's page, or can I update one beer but add the complete list in the dialog box for that first beer?

 

Personally, I think this is a good decision. There are several small breweries around me that just don't get the rating "foot traffic" on BA that they get on Untappd and their beers were retired even though they are regularly on tap or rotate seasonally.

 

Thank you. I have reviewed quite a few up here in Oregon that were supposedly retired.
We don't tend to review enough as a group.

 

This was my biggest gripe. My area doesn't experience a lot of BA traffic in general, with only the most popular breweries achieving a BA score on a handful of beers (after years of releases, mind you). The auto-retiring feature felt like a kick in the caboose without the local site traffic to keep the site accurate. Lots of false 'retired' beers. Happy to hear changes are being made as I think it'll help my area.

 

I think it helps BA to have those listed too. I've noticed the majority of the time when I do google search on a beer name the BA reference comes up in the top 3 to 5. That's good exposure for BA.

 

Many of these types of issues are ultimately spawned in large part by having a small user base in many locales. Never thought I'd say this, but this could be a use case where it's worth a "soft," behind-the-scenes, and one-way integration of the BA and Untappd databases. It doesn't have to be perfect, and it doesn't have to impact beer listings or scores, but BA could query the last Untappd check-in for a beer of the same name from the same brewery before classifying it as active or inactive. I'd go with a fuzzy scheme where every word in the BA beer name has to be in the Untappd name, and some words can be substitutes (e.g., "BA" and "barrel-aged" and "barrel aged" are equivalent), to deal with non-standardization in how names get formatted, but that would likely be a major improvement that remains scalable without relying on user input.

 

Thanks. I figured that. A brewer's request to specifically "retire" a beer would make sense if BA didn't make any distinctions or if BA had a "retired" distinction. If retired was replaced with "inactive," then brewers can request that a beer be moved to "inactive." It serves a similar purpose but uses a different label. I can't imagine brewers being too miffed if such beers were called "inactive."

With a non-automated "retired" system, if Sierra Nevada decides to bring back regular Otra Vez as a temporary taproom draught, or as one of their very small can runs, or as a beer in a mixed pack - either BA site users will have to live with the beer being identified as "retired" when it isn't (and if that doesn't matter, then I'm not sure how much any of this matters)... or BA mods will have to switch it to active (perhaps after a user report) and then remember to switch it back to retired at some point (I assume).

Given the high volume and degree of fluidity with beer releases today... it seems to me that having a manually regulated beer retirement system is not as critical or relevant as simply having a beer listed and represented in the database. Those are obviously not mutually exclusive, but with a finite amount of time that one can spend keeping the database up-to-date, it might be better to use that bandwidth to work on the addition of beers rather than the maintenance of what's already here.

None of this is meant as a criticism... it's just food for thought in case an additional perspective helps. I clearly value your site. I suspect users would be annoyed if the site took on the approach in my posts.

 

I don’t agree with this. In the otra vez example, which is a really good example, hopefully it gets flagged as not retired (I acknowledge the possibility it might not, but it’s more likely than some other breweries), then it would be shown on the regular list until it gets specifically marked as retired. I don’t think it should be retired when the draft kicks at SN; they’ve shown that they will occasionally produce the beer at that point. And the current system will work that way.

Doing it the way you suggested - via automation - if someone checks in a Don quijote tomorrow - now it will be listed as active on the BA250 for 2 years.

That particular beer may be an extreme example, but there are more beers that would fall into that grouping than you’d think. Take BCS proprietors - around with a different flavored vintage since 2013 - I just checked and the only one that wouldn’t be active if you went by checkin recency is Prop 2015, which is the infected year (and still almost makes the cut). A bunch of those beers would make the BA 250, and practically all would make the Illinois 100. In my opinion, that would render those lists meaningless.

 

I like the automation aspect, because I see so many beers on here that I find fresh incorrectly listed as "retired". But also see @Chuckdiesel24 point. What if you kept the automation to unretire beers, but increased the requirement for beers to make the leaderboard with a cutoff, say 5 or 10 or some number of reviews within the past n number of years? Or maybe if the beer hits that threshold of recent reviews, it trigger a BA moderator to review the beer as active? Again, just ideas here.