Take the 2-minute tour ×
Homebrewing Stack Exchange is a question and answer site for dedicated home brewers and serious enthusiasts. It's 100% free, no registration required.

So, I am very new to home brewing, my first batch has been fermenting for about a week. I have been a beer enthusiast for years, so I am really enthusiastic about this new hobby of mine. By trade, I am a computer scientist and I specialize in data analytics. I use a tool called QlikView, to build analytical applications, and it is really powerful in spotting relationships between data.

So, I have an idea to build up a database (really, just a series of Excel Spreadsheets) and then load them into a QlikView application to allow us home brewers to do some analysis on the composition of existing beers, to allow us to "tinker" with a recipe based on ingredients that are shared in beers we like.

An example I can think of... say I have three favorite IPAs, if I saw that they all share a common hop then I would view this as a hop I must really like in a beers. The idea would be to try to build up new recipes based on common components across potentially different clusters of beers someone may like. This could extend beyond just the ingredients, also into the brewing process and additions, etc.

So, my question is, has anyone tried something like this? (obviously, not necessarily in the tool I am suggesting). Also, is there a good repository of beer recipes out there I could use to seed this database? Also, would anyone be interested in joining in such a project?

No financial angle here, I am wanting to build something useful that would be openly shared.

Thoughts?

Chris

share|improve this question
    
I'm a software guy too (no surprise given that we're on the Stack Exchange site network right now). The subjectivity of taste comes into effect here, making your idea a little tricky. Also, we're talking about a LOT of factors (hops, hop interaction, malts, yeasts, temperature, mash diff) across a SMALL sample size (most brewers prob average 10-30 batches per year). Not to say that your idea can't be done, but it will be tough. –  Graham Dec 23 '13 at 13:51
    
I've thought about building an online database of single ingredients based on their flavor characteristics, allowing for users to enter flavors they desire or dislike, and to receive suggestions for how well certain ingredients match those flavor. So entering "citrusy" would return Cascade or Centennial hops as a strong match, something like Glacier as a weak match, and Hallertaur as a negative match. Its a simpler version of the system you are talking about. –  Graham Dec 23 '13 at 13:55
    
@Graham Check out Hop Union's hop aroma wheel as an example. –  Chino Brews Dec 24 '13 at 21:29
add comment

2 Answers

In my opinion, metrics and analytics are always important, and useful. The problem is pitching your idea to a broad enough audience to get it off the ground. Without mass data to back everything up, you won't be able to justify the work. If you intend to do it for yourself only, well I honestly wouldn't bother, since I know what I like, and what I don't like. I wouldn't need a spreadsheet to verify it.

You say you are a computer scientist? There's repositories packed to the brim with recipes, and they may be willing to work with you, if you are willing to get your hands dirty and bust out your coding chops. I would recommend contacting the folks at Brewtoad, or Brad on Beersmith and pitch your ideas to them.

To answer your question, on a homebrew level, no. I am unaware of any site or tool that parses through data to show trends and analysis for ingredients and techniques. A lot of it is subjective, since it really does boil down to (ba-dum-tiss) flavor, which is always subjective to each person. I would guess there's hesitation from developers since there really is no right or wrong way to brew a beer. You would have to keep everything factual.

share|improve this answer
    
Thanks for the leads about Brewtoad and Beersmith! If I decide to get this off the ground I will reach out to them. I understand your point about knowing what you like, and I feel the same way, however I would be curious if there are beers out there that I may not even think I would like, and by spotting connections between it and a beer I know I like, I could expand my tastes! –  Nibroc A Rehpotsirhc Dec 22 '13 at 6:42
    
Alright... so I signed up on Brewtoad, and I have spent 5 minutes and I am thinking this site was exactly what I was looking for! Thanks again for sharing. –  Nibroc A Rehpotsirhc Dec 22 '13 at 6:53
    
You're welcome. Brewtoad has active developers that like to push out updates often. While a great repository for information and recipes, it also lacks quite a bit in statistics, analytics, and they don't even have something as simple as a rating system. I'd suggest reaching out to them. I've talked with Brad from Beersmith, and he likes to run a tight ship, you may not find much success there. –  Scott Dec 22 '13 at 18:24
add comment

I don't have enough reputation to add a comment, so I'll just post an answer to participate.

This sounds sort of like a discovery service for beer, kind of like Last fm and Pandora are for music, correct? Although it seems like there would be more detailed match criteria with the ingredients included. After a quick Google search, I can only find a few (BrewGene, RateBeer).

If I am understanding you correctly, you would have the following tables: brewery, beer, hops, yeast, fermentables and link tables for many to many relationships with maybe a few more tables for hop addition time, other ingredients, etc. Basically you'd have a recipe database for each beer so that relationships can be determined using the likeness of beers (beer A is similar to beer B in terms of the hop schedule or grain bill, etc.).

Yes, you would have to find a good repository of beer recipes to work from, otherwise the research would obviously be a lot of work. There would also probably a degree of completeness/correctness of these recipes to consider. Maybe the users can submit the ingredients/recipes and suggest changes to existing recipes as a collaborative effort sort of like Wikipedia? As far as existing recipe databases go, all I can find are homebrew recipes rather than commercial ones.

I am also a computer scientist specializing in web application development. I've never used QlikView, but looks interesting.

share|improve this answer
    
Yes, this is exactly what my idea would be. Good idea about the correctness of the recipe, as I have to imagine that some of the home brew recipes circulating on the web are less than accurate. What web development stack do you prefer? Perhaps we can collaborate. –  Nibroc A Rehpotsirhc Dec 22 '13 at 6:39
    
Looks like Brewtoad has everything covered? That's a shame because this would be fun to work on. Most of my recent experience is with ASP.NET/SQL Server, but can work with anything else. –  jedi jay Dec 23 '13 at 17:48
    
BrewToad is certainly impressive, but the more I used it I still think there could be room for improvement. It focuses more on home brew recipes, not commercial recipes. In addition, it doesn't have the same "associative experience" I was imagining with QlikView, or one that I imagine could be pulled off in a native ASP site. I am a SQL Server guy, I can code .Net, know LINQ, but I am useless with CSS and making a website look pretty. I have dabbled in MVC as well. I am still game to try to put something together. –  Nibroc A Rehpotsirhc Dec 24 '13 at 3:48
    
I have done a lot of work with building a front end with JavaScript and CSS... not an expert, but still have a lot of confidence. Shall we take a stab at it? –  jedi jay Dec 27 '13 at 2:31
    
I'm game, shot me an email. Christopher.a.corbin@outlook.com –  Nibroc A Rehpotsirhc Dec 29 '13 at 5:46
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.