The on-wiki version of this newsletter is here:
https://meta.wikimedia.org/wiki/Abstract_Wikipedia/Updates/2021-01-21
This week’s newsletter is a bit more technical than others. Before we get
to the main content, a reminder: the logo concept submission
<https://meta.wikimedia.org/wiki/Abstract_Wikipedia/Wikifunctions_logo_conce…>
is currently open! Six logo proposals have already been made, and they are
each very much worth a look. I hope to see more proposals coming in, the
submission deadline is on 16 February.
The formal function model
<https://meta.wikimedia.org/wiki/Abstract_Wikipedia/Function_model> as it
is currently written assumes that the whole of the function model is
already implemented. In particular, it relies on generic types, which means
a type that is parameterized, usually by another type.
What is an example of a parametric type? Let’s take a look at the type for
lists. A list is an ordered series of elements. What is the type of these
elements? It could be anything! There are plenty of operations on lists
that can be performed without needing to know what the type of the elements
in the list is. One can take the first element of the list, or reverse the
order of the elements in the list, or take every second element from the
list, or much more.
What is the type of the item that is returned by the function that outputs
the first element of such a list? You can’t know. Since the elements of the
list can be of any type, the return type of the function outputting the
first element could also be any type.
Now, instead of having a single type called "list" you could also have
types called, say, "list of strings", or "list of numbers", or "list of
booleans". Or, even more complicated, "list of lists of strings". You could
also have a list of any kind of elements, or multiple kinds at once. Now,
if you have a function that returns the first element of a "list of
strings", you know that the return type of that function will always be a
string. You have more type safety
<https://en.wikipedia.org/wiki/Type_safety>, and you can have your
code-writing tools provide much better guidance when writing functions and
function calls, because you know that there must be a string. It is also
easier to write the functions because you don’t have to check for cases
where there are elements of other types popping up.
But on the other side, we suddenly need many more new types. In theory, an
infinite number of specialised types. And for each of these types, we would
need all the specialised functions dealing with that type, leading to an
explosion in the number of functions. This would be a maintenance
nightmare, because there would be so much code that needs to be checked and
written, and it all is very similar.
In order to avoid that, we could write functions that create the type and
functions that we need, and that take the type of the elements of the list
as an argument. So instead of having a type “list of strings”, we would
have a function, call it “list”, that has a single argument, and that you
can call “list(string)” and that would result, on the fly, in a type that
is a list whose elements are strings. Or you can call “list(integer)” and
you get a list whose elements are integers.
The same is true for functions: instead of having a function “first” for
that only works on lists of string and returns a string, you would have a
function that takes the type “string” as an argument and returns a function
that works on a list of strings and returns a string. Instead of writing a
“first” function for every kind of list, we would write a function that
creates these functions and then call them when needed.
There are also situations where the dimension of typed input on which you
need to operate is limited by number, rather than or as well as type. As a
more complicated example, if we had a method to do matrix dot multiplication
<https://en.wikipedia.org/wiki/Matrix_multiplication>, the number of
columns in the first input matrix and rows in the second must match. There,
instead of taking just a type to create the matrix (say, a floating point
number), our top-level type function would take two integers as the numbers
of rows and columns as well. We could then call this method with
matrix(float,4,3).
Similarly, Earth-based ground positioning information is generally relayed
by two dimensions of degree, and optionally one of altitude; there you
might have tuple(float,2) for the former, and tuple(float,3) for the
latter. The exact way that the Wikifunctions community would decide to
model the position type is left up to them – in deciding on a type, they
might also want to explicitly specify the planetary body, the datum, the
accuracy in each dimension, or other data as well. We just need to make
sure we provide the flexibility to editors to represent things they will
need or want. Note also that common types, such as geocoordinates, will
likely be created as named types on wiki, but structures can always be
created on-the-fly too.
This idea has many different names and concrete realisations in different
programming languages, such as templates, concepts, parametric
polymorphism, generic functions, and data types, etc. For our purposes, we
call this generics, and it is currently scheduled to be implemented in Phase
ζ
<https://meta.wikimedia.org/wiki/Abstract_Wikipedia/Phases#Phase_%CE%B6_(zet…>.
Now the thing is that the function model as it is currently described
relies heavily on generics. But until they are in place, we can’t really
use them. So we are in kind of a limbo, where the precise function model we
are implementing right now is not specified anywhere, and instead we are
adjusting on the fly based on the current state of the implementation and
where we want to eventually end up.
In order to support that, we are publishing an explicit pre-generic
function model
<https://meta.wikimedia.org/wiki/Abstract_Wikipedia/Pre-generic_function_mod…>.
Note that this also does not describe the model as it is right now, but it
is the model that we are going to implement mostly by the end of Phase γ
<https://meta.wikimedia.org/wiki/Abstract_Wikipedia/Phases#Phase_%CE%B3_(gam…>
and thus is much more immediately useful than the final function model that
we have currently described on-wiki. The idea is then that once we get
around to support generics, we will shift over from the pre-generic
function model to the full function model.
Comments and suggestions on the pre-generic function model and the plan
presented here are, as always, very welcome.
Background Wikipedia articles:
-
Generic programming <https://en.wikipedia.org/wiki/Generic_programming>
-
Parametric polymorphism
<https://en.wikipedia.org/wiki/Parametric_polymorphism>
-
Generic function <https://en.wikipedia.org/wiki/Generic_function>
-
Type safety <https://en.wikipedia.org/wiki/Type_safety>
And remember the ongoing logo concept proposal contest
<https://meta.wikimedia.org/wiki/Abstract_Wikipedia/Wikifunctions_logo_conce…>
!
The on-wiki version is here:
https://meta.wikimedia.org/wiki/Abstract_Wikipedia/Updates/2021-01-28
Today I want to give an update on the development work, with a particular
highlight on the community contributions - and renew our invitation for
volunteer developers to join!
We have defined the work to get us to the initial launch of Wikifunctions
into 11 phases, of which we have finished one. Phase α
<https://meta.wikimedia.org/wiki/Abstract_Wikipedia/Phases#Phase_%CE%B1_(alp…>
was about the ability to create objects on wiki, and our current Phase β
<https://meta.wikimedia.org/wiki/Abstract_Wikipedia/Phases#Phase_%CE%B2_(bet…>
is about creating new Types and creating instances of such Types.
In the backend of the wiki, we have been mostly busy with turning the
hard-coded validators we had so far into validators that are based on the
type definitions inside the wiki. This task is still on-going, and will
remain so beyond Phase β.In the end, the validation of instances will be
performed by functions written by the community and stored on the wiki
itself, so we will need the ability to define and run functions in order to
fully allow their validation. Until then, we will have incomplete
validators that will become increasingly more comprehensive.
The frontend work for supporting user-defined types has also progressed,
and here we saw a number of contributions from two volunteers, Gabriel Lee
and Arthur P. Smith. Thanks to them, a lot of the identifiers are now also
enriched with the labels in the user’s language, and adding data is
becoming much easier than before.
As usual, you can see and experience the results right on the "notwikilambda"
demo system <https://notwikilambda.toolforge.org/wiki/Main_Page> set up by
another person volunteering, Lucas Werkmeister, who in their "day job"
works on Wikidata.
We try to reach out to a volunteer when we notice them contributing code.
We plan to invite them to chats with individual members of the team,
discussing the tasks, and also invite them to some of our daily stand-up
meetings, if the timing permits it. We have had a few people join one or
more of our stand-ups so far, and we are looking forward to inviting more.
There are several ways a developer can contribute at the given time, and
the ones that I am going to list here are by far not complete.
One way is to take a look at our task board on Phabricator
<https://phabricator.wikimedia.org/project/view/4876/>, and see if there is
a task that you would like to work on. This is probably the best way to
contribute to the WikiLambda extension
<https://www.mediawiki.org/wiki/Extension:WikiLambda> itself, to work on
our UI front-end and the MediaWiki back-end. You might instead find a task
to work on within the stand-alone services that will support Wikifunctions
– the function-orchestrator
<https://gerrit.wikimedia.org/g/mediawiki/services/function-orchestrator>
and the function-evaluator
<https://gerrit.wikimedia.org/g/mediawiki/services/function-evaluator>.
Beyond Phabricator, there are also development projects which are less tied
in to our daily development work, and I would love to see those happen. For
example, Lucas Werkmeister has started an alternative implementation of the
function evaluator <https://github.com/lucaswerkmeister/graaleneyj> of an
earlier prototype of the project, based on GraalVM. Other, parallel
evaluation engines would also be very interesting. For example, one running
on cloud computing resources, one running in the browser, or one running on
the local machine.
Besides that, I would love to see alternative interfaces to interact with
the Wikifunctions system once it is in place. Possible interfaces might be
viewing and/or editing via a modern CLI, through a Website that is not
based on MediaWiki (perhaps hosted on Cloud Services
<https://wikitech.wikimedia.org/wiki/Help:Cloud_Services_Introduction>),
through a voice-based or touch-based interface, within a Hadoop workflow,
or even within a spreadsheet. Some of these concepts might still be a bit
premature, but if you are interested in one of these, let me know and we
will find a place on-wiki to collect and discuss ideas.
I would particularly love to see an interactive visualization of a function
evaluation, based on a composed implementation, allowing the user to better
understand how a specific value is derived. Such projects can be started
more or less independently of the main Wikifunctions work (although,
admittedly, it might often make sense to wait until the project has
stabilized a bit, in order not to develop against a moving target).
As the project progresses, we expect that more and more people with an
increasingly diverse skill set will feel able to contribute to the project.
Until then, there are already a few other ways you can contribute if you’re
not a developer:
-
We are currently looking for submissions of logo concepts
<https://meta.wikimedia.org/wiki/Abstract_Wikipedia/Wikifunctions_logo_conce…>
(we also moved the submission deadline for logo concepts to 23 February.
-
You can join the Wikidata community to increase the coverage of the
lexicographic
knowledge in Wikidata
<https://www.wikidata.org/wiki/Wikidata_talk:Lexicographical_data> (we
are working on a few ideas in this direction that we will talk about in the
coming weeks).
-
You can read through our documentation
<https://meta.wikimedia.org/wiki/Abstract_Wikipedia> and ask questions,
make clarifications, and help us to make the project more accessible.
-
There was the suggestion to collect possible user stories
<https://meta.wikimedia.org/wiki/Abstract_Wikipedia/User_stories>, which
would also be very welcome.
As the project progresses, more and more tasks will become available. We
hope you will keep an eye on our progress, and join us whenever you feel
ready to contribute!
The on-wiki version of this newsletter can be found here:
https://meta.wikimedia.org/wiki/Abstract_Wikipedia/Updates/2021-01-15
We are looking for the Wikifunctions logo concept
<https://meta.wikimedia.org/wiki/Abstract_Wikipedia/Wikifunctions_logo_conce…>
!
Just as with the name, we are starting with a request to submit proposals.
As time of this writing, three proposals are already in - we are looking
for more proposals for logo concepts. The deadline for submitting proposals
is in a month from now, February 16, 2021, followed by a single round of
voting.
As with Wikidata, the proposals don't have to be fully polished. There will
be a final step of legal review and design polishing. We invite everyone to
participate in the logo concept process and submit your proposals.
I am amazed by the creativity and productivity of the communities to come
up with logos for our projects. My personal favorite, by quite a stretch,
is the Wikidata logo. Maybe it is because I am obviously biased towards
that project, but I also love its recognizability, the layers of meaning it
has, its flexibility to change and accommodate events and situations by
changing sneakily the word that the bar encodes, etc.
I wish that the Wikifunctions logo will contain a similar form of
flexibility. That it is similarly recognizable, even when sized down a lot
(the favicon for Wikidata
<https://commons.wikimedia.org/wiki/File:Wikidata_Favicon_color.svg> is
really neat).
I could imagine that going through the works and notes of people like Alan
Turing, Ada Lovelace, Gottfried Wilhelm Leibniz, could lead to interesting
inspirations. A Turing machine? Lambda calculus? Aristotelian logic and its
various notations?
Maybe something more practical. A punched card as it was used for the
Jacquard loom? A punchhole card could make an interesting basis for a logo.
Gears and steam-power are another possible basis, as they have implemented
‘smart’ appliances such as automatic doors for millenia. Or semiconductor
chips and boards.
All of these have the potential to look very detailed and fizzy when they
get scaled down. So it might be worthwhile to explore slightly different
designs at different sizes. E.g. the number of holes in a punchhole card,
or the number of circuits on a board might be different. For example I
could imagine a chip in the center with a lambda embossed, where circuits
connect it to the other Wikimedia projects, with highly stylized logos -
just a rectangle for Wikidata, a triangle for Wikivoyage, nine little
rectangles for Wiktionary, etc. And the smaller version is just the lambda
with lines going out, etc.
There is no need to follow the Wikimedia colors. But there is also no need
not to. I think the incorporation of these colors in the Wikidata logo
worked out beautifully. But the logo of Wikisource is no worse due to not
incorporating the colors.
One proposal is already on wiki. I am adding my own proposal, because it is
so easy to create: a backslash, a lambda, and a slash, looking like a W in
many fonts: \λ/ [image: 😉]
<https://meta.wikimedia.org/wiki/File:Emojione_1F609.svg>
I am looking forward to your ideas, and just want to start the conversation
and the brainstorming here. The goal is to have discussions and ideas, and
everyone should feel comfortable to borrow from each other in creating
proposals for the first round. Let’s discuss possible inspirations and
ideas for the next few weeks. I would love this to be much less a
competition or a contest with a single winner, and much more a
collaboration and cooperation, where, in the end, we all win.
Here are also some great words of consideration by Zack McCune
<https://meta.wikimedia.org/wiki/User:ZMcCune_(WMF)>:
Logos are visual tools. A logo uses recurring graphic elements (made of
colors, shapes, icons) to represent a company, initiative, project, or
organization without any words.
When used in combination with a formal name, the precise arrangement of the
logo and name are called a brand lock-up. (Here’s a handy set of examples
from the University of Indiana in the United States
https://brand.iu.edu/design/logos-lockups/marketing-lockups.html)
Making a great logo is about balancing a few graphic design factors:
1. Recognition - Can people understand what is depicted?
2. Association - Do the graphic elements communicate the project or
company’s purpose? Do they link the project or company to a family of
related brands? Does the design suggest connection to a theme, object, or
process essential to the project or company?
3. Originality - Is the graphic unique enough to not be confused with
existing projects or companies?
4. Versatility - Will the graphic work BIG and small? Can people use the
graphic easily (e.g. place it correctly in new designs, adapt it for wide
application), and use it quickly possibly even drawing or approximating it?
In the Wikimedia world, association is one of the most important qualities
for a new project logo. There is much recognition (and love, which
marketers call “affinity”) for Wikipedia and Wikimedia projects and
referencing graphic elements (colors, shapes, symbols) from these existing
logos will help BUILD upon their reputations.
As you look to make a logo, TRY LOTS OF APPROACHES. Collect symbols
associated with the topic. Look for patterns/trends in the symbol sets.
Universities, for example, often using heraldic design traditions
<https://www.universityaffairs.ca/features/feature-article/the-coat-of-arms-…>
to
suggest history and credibility, but are also criticized as out-dated and
overly western. Consider where you want to be original and where you want
to use associations. Noun Project
<https://en.wikipedia.org/wiki/en:The_Noun_Project> searches can be helpful
for finding symbols linked to phrases or topics.
Logos also have stories embedded in them. Consider how the artwork you are
making will be part of future conversations. The Wikidata logo, for
example, intentionally resembles a bar code. This is quickly apparent and
communicates the project’s purpose. But there is more to learn about the
logo… and that story deepens the care and joy people have for this brand.
Simplicity in design is hard. Sketch many many ideas! Try to limit yourself
to a small drawing area or with a fixed stroke width to make ideas even
more minimal. Also: do not *start* with color. Logos are registered in
black and white and must function as memorable single-color expressions to
be effective across many use cases (e.g. on t-shirts, on slides, on
multi-colored backgrounds).
Got a set of logo ideas? Great! You should also TEST logo ideas with
friends, family, colleagues, and *ideally* the audience of people you WANT
to have a functional/emotional reaction to the design. Feedback will help
ideas improve and grow towards highly memorable and effective graphic
design. Further design guidance now at the Logo page
<https://meta.wikimedia.org/wiki/Logo> is adapted from the great advice
originally captured
<https://www.mediawiki.org/w/index.php?title=Project:Proposal_for_changing_l…>
by
my colleague Volker <https://en.wikipedia.org/wiki/User:Volker_E.>.
The process will take a while, and we explicitly reserve the freedom to
modify it as we go if we see that things are not working out as planned.
And our first step is to come up with a sizeable number of ideas and
proposals, before we start to vote and whittle them down to a small set of
core ideas that in turn can be expanded upon.
Join the Wikifunctions logo competition
<https://meta.wikimedia.org/wiki/Abstract_Wikipedia/Wikifunctions_logo_conce…>
!
I am looking forward to what you come up with!
The on-wiki version of this newsletter is here:
https://meta.wikimedia.org/wiki/Abstract_Wikipedia/Updates/2021-01-07
We are very happy to welcome Cory Massaro, who is joining the Abstract
Wikipedia team as a software engineer.
I will let Cory introduce himself in his own words:
“I obtained my B.A. in creative writing and classical languages in 2010,
then studied comparative literature with a focus on oral history and epic
before switching to computational linguistics. I then spent several years
working on speech recognition systems and internationalization. More
recently, I’ve taken on mentorship and educational roles with a particular
focus on helping entry-level workers build class consciousness.
“My core passions are teaching/learning, the arts, language, and the
relationship of these to workers’ rights. If there is a throughline uniting
these interests, it is this: I care deeply about knowledge. I want to
contribute to a culture where people possess sufficient education, leisure,
and resources to relate to knowledge and act on it. To these ends, I have
contributed research analyzing how ideology can create dialect. I also
conduct workshops on political rhetoric. In my free time, I write, learn
languages, create text adventures, and play music.
“I discovered programming as a matter of necessity. Many career paths
seemed unreachable due to student loans, while the for-profit tech sector
offered a path to stability. Only later did it become clear that
values-driven tech work existed.
“For this reason, I’m delighted to be joining an organization built on
principles of freedom, openness, and cultural/linguistic pluralism. I can’t
wait to help make knowledge more accessible and am excited to learn more
about this fundamentally democratic approach to language and community
engagement. Plus, if there is a modern global analog to the ancient keepers
of oral history, Wikimedia is probably it, and that is beautiful.”
Please, join me in welcoming Cory to the team!
Also, as you may already know, in December we finalised the name for the
new Wikimedia wiki. The name for the repository of algorithms will be
Wikifunctions, as voted for by the community. The domain name
wikifunctions.org has been secured.
The next step will be to start the process to select the logo concept for
Wikifunctions, which is slowly kicking up its speed. Submissions can be
made here <https://meta.wikimedia.org/wiki/Abstract_Wikipedia/Logo>, and in
the following days we will publish a few supporting documents and re-do the
timeline for the process in order to make it a bit simpler. This is similar
to the approach used for the Wikidata logo. I am very much looking forward
to seeing the process play out!
