Models For Biology User Types (Personas)?

Question

[Egon Willighagen]

I blogged today about usability, noting that usability is a multi-objective optimization problem. The multiple objectives are defined by having multiple user types (the proper term is persona) in biology. We have, without any judgement, Excel users, R hackers, workflow designers, the absolute novice student, etc, etc. You all know them, as they are all present on this forum.

Each user type has their own set of requirements: good looks, a clean API, etc. All at various levels: it must support .xsl files; we need SOAP access; it must comply to our minimal reporting standards. Etc, etc.

Now, I was wondering if some templates have been developed defining a number of common user types in biology. I am unaware of such in chemistry. Do such models exist already? Where have these been published?

Just as an example of what we should be thinking about it this post by Lars Wirzenius. While he talks about roles rather than user types, it does exactly what we should be doing: identify users or user roles to see how these map to requirements. In this post he identifies a new user type, with a different role. Lars lists these roles:

• developers
• sysadmins
• users

These come from the Unix/mainframe model. User types I am thinking of (user types can have multiple roles, and one role can be performed in different ways, and thus with different requirements, by different user types):

• bench biologist
• professor
• data analyst
• bioinformatic software developer

They will all have different requirements for a certain joined platform.

[Pierre Lindenbaum]

There must be a xkcd for this question isn't it ? http://xkcd.com/927/ :-)

[Egon Willighagen]

This is why I ask... I hardly feel like defining new types... who all know who they are. I'd rather pick something from the shelve, and then continue with the things I'm really interested in...

[Pierre Lindenbaum]

Egon, I'm not sure about what you're looking for. Are you looking for something like the IAO http://code.google.com/p/information-artifact-ontology/ (It's an ontology but I don't think it is widely used )

[Egon Willighagen]

IAO is actually used in our CHEMINF ontology :) But I don't think IAO can be used to capture user types or user roles...

[Egon Willighagen]

Pierre, I give some practical examples of what I am getting at: a role -based model (existing model), and a type-based model (rough idea).

[Pierre Lindenbaum]

Ah , I see, you want something like a user-type-ontology

[Pierre Lindenbaum]

Something like the geek-code ? http://www.geekcode.com/geek.html

[Egon Willighagen]

No, I want literature on where they created such an ontology.

[Egon Willighagen]

But then for biology specific.

[Egon Willighagen]

Pierre, they are a bit like PhD characters: http://www.phdcomics.com/comics/aboutcomics.html but then focusing more on the work they are expected to do, rather than the live around it.

Answer 1

[Søren Mønsted]

Michael made me aware of your post. I am working with user experience at CLC bio (a professional bioinformatics software company) and we have used personas in our development process.

Our personas were created to make our users real and present in the team of developers (who are quite technically minded). I think their nature is less descriptive and more communicative compared to the user types you are writing about. It's been a while since I have worked with these personas, but if you want to have a look, just send me an email and I will try to find some pdfs for you ([e-mail]).

Answer 2

[Michael]

The most developed user type categorization scheme I've seen so far is used by CLC Bio in Denmark. It was developed by someone with formal usability background who brought this into their software development early on. In my own experience there are at least 3 common user types: a) the typical wet lab biologist who is used to MS Office and a few copy/paste bioinformatics web tools, b) the full-time bioinformatician with extensive scripting/programming and data analysis skills, c) a user who is somewhere between both (e.g. a biologist with above-average bioinformatics skills who spends a considerable amount of time on data analysis). All 3 have vastly different requirements, and knowing one doesn't help you infer what the others need.

[Mary]

Oh, yeah--I remember talking with them about that once, but didn't see the full strategy. Good one.

Answer 3

[Mary]

I have a friend who was very interested in usability and developed user types categorizations like this. But I don't think it was ever published. I can ask him. I'll edit this if I get an answer. He was a real evangelist for usability--but it was never really taken very seriously by the projects. That always made me sad....Alas.

Today I was also looking at this paper that set up different profiles for different users to evaluate the software. http://www.biomedcentral.com/1756-0500/4/133 If you go into their database you can see some of the criteria that have been established for "biologist" vs "computational biologist" for example.

Answer 4

[bryan.smith4756]

It sounds like you're looking for templates or models that define different user types in biology. While I don't know of any specific templates for biology, the idea of understanding different user types is important for creating user-friendly tools or platforms. For example, in a post by Lars Wirzenius, he talks about roles like developers, sysadmins, and users in the Unix/mainframe model. In biology, you can think of user types like bench biologists, professors, data analysts, and bioinformatic software developers, each with their own unique requirements. Unfortunately, I'm not aware of any specific models or publications on this topic in biology, but you can explore related research or publications in the field to gather more insights.