Microsoft ART team Job Position

mike_duggan's picture

A job opening has become available with the Advanced Reading Team at Microsoft.

Here is a link to the official job posting

https://careers.microsoft.com/jobdetails.aspx?ss=&pg=0&so=&rw=1&jid=7326...

cheers
Mike

k.l.'s picture

– "superstar"
– "previous experience on the future of reading is recommended"

Is this really the official one? ;-)

– "to make reading on screen a better experience than reading from a book"

Make high-res screens.

John Hudson's picture

Make high-res screens.

That seems a sound approach to eventually making reading on screen as good an experience as a book, but I think to make reading on screen a better experience you need to start thinking about what you might be able to do with text on screen that you can't do in a book.

The job description is pretty funny -- especially the previous experience of the future --, but since I know the person whom they're trying to replace it makes some sense to me.

Karl Stange's picture

"but I think to make reading on screen a better experience you need to start thinking about what you might be able to do with text on screen that you can't do in a book."

This is one of the major problems facing traditional publishers. There are many talented designers working on print material that do not understand how to engage with digital technologies beyond trying to transition traditional content to new formats and platforms. It is difficult but I think there is a lot to be gained from getting those designers to think about and work natively in digital publishing environments. Typographically so much more is possible now than even a couple of years ago and the technology is developing all the time.

The Future of Reading session at the recent Tools of Change conference was a good overview of what is possible but most publishers are only just getting to grips with these concepts, let alone the tools that make great digital typography possible.

It sounds like a fascinating job and it is great that Microsoft continue to see the importance of those elements as part of single focused role.

k.l.'s picture

J.H. – but I think to make reading on screen a better experience you need to start thinking about what you might be able to do with text on screen that you can't do in a book

I agree. At the same time MS's "better" differs from your "better": The job description, in light of past activity, is mostly about the medium side, dealing with screens and their possibilities and limitations as to what they can display. What you bring up now is the narrative or storytelling side, which is not something I saw MS being concerned about. (And they shouldn't or don't need to.)

K.S. – was a good overview of what is possible ...

The aspect of adaptive/flexible (and typographically refined) layout as made possible by wpf or html5+css3 is an old hat to me. Still I have trouble to understand why it did not happen ten years earlier.
Taking into account what John said and approaching it from the storytelling side, the adaptive-yet-typographic-layout technology is means rather than ends, it is but one core material condition among others to make new kinds of storytelling possible.

K.S. – ... but most publishers are only just getting to grips with these concepts, let alone the tools that make great digital typography possible

It needs new generations of publishers and designers.
Current print designers have trouble enough getting the idea of adaptive layout, which in turn requires conceptual/structural thinking rather than the static collage type of thinking. Ironically enough, an earlier generation of book typographers, working with tools like Berthold typesetting programs, were familiar with structural thinking – but most likely are long retired. (Put differently: PageMaker, XPress, InDesign, following a fixed collage mode of thinking, caused typesetters/typographers to un-learn structural thinking.)
And I think you over-emphasize the aspect of learning tools. Tools are easy to learn when you know what you want to do with them. The problem is that designers/publishers don't know what they want.
Moreover, what's an ebook, since you pointed at the presentation? Isn't that part of the problem rather than being the solution? Mostly it is about packaging old content in a slightly modified way ...

Kevin Larson's picture

We've changed the job title to Leading Type Expert.

Nick Shinn's picture

Speaking of advanced reading technologies, I have a new pair of glasses which I occasionally use to decipher handwriting.

But seriously:
I won’t be applying for the job, however, there is one thing I think the ART team really could do, and that is to incorporate readability measurements into the type design process, i.e. product development. This would involve making variants of a typeface and testing which reads best. So far, all I have ever come across in the science of reading is one published typeface measured against another quite different typeface, or at best a face refined with ad hoc feedback from users. I’ve designed commissioned faces where several variants have been presented to the client, and these have been tested in mock-ups with focus groups, but the feedback is ad hoc rather than scientific.

eliason's picture

We've changed the job title to Leading Type Expert.

So somebody really good at linespacing then? ;-)

John Hudson's picture

Karsten, 'storytelling' -- new forms of literature -- strikes me as only one area in which reading on screen might differ from books, and not personally interesting to me. I spoke specifically about what one might be able to do differently with text on screen, and I meant in the visual sense of text not the literary. Sorry if that wasn't clearer. One obvious thing is the use of colour, which has tended to be absent from typographic printing -- due to cost, I think -- but which has a rich manuscript tradition, e.g. in Arabic and some Jewish vocalised texts and in Ethiopic in which bichromatic punctuation signs are normative in the manuscript tradition. The ART team has been exploring ways in which empirical testing might feed back into design -- which is why when they talk about 'better' they usually have an actual measure in mind --, and this strikes me as an area in which one might be able to determine whether something like traditional use of colour in text might actually aid aspects of reading such as comprehension. And if it does, then it raises the question how else might colour be used, and what would the result be?

Printing has tended to be conservative; more so than the manuscript tradition in many respects. In large part this seems to me to have been the result of the manufacturing costs of type for most of the past 550 years, and the the material costs of publishing. No one wants to take risks when so much is at stake. Publishers are rightly wary of publishing books using novel types, let alone introducing new punctuation symbols or varying conventional typographic norms. What if no one wants to buy it? But the same electronic document can be read on screen in any variety of style sheets, and the reader can even apply his or her own preferences. How might this be extended? Style sheets for the most part remain static: they present a single way of viewing and reading, selected by the publisher or, less often, by the reader. Are more dynamic methods possible? Part of the job of a typographer is to select a typeface and layout appropriate to the subject matter and literary style of a document, taking into account both convention and possibilities for innovation. We all have an idea of what a novel looks like versus an annual report, and of what range of typeface options might apply to each. My guess is that a computer can very easily be taught to recognise what kind of document a reader is looking at, so could apply appropriate styles automatically. Or a reader could say 'This is what I want novels to look like when I read them' and 'This is how I want to read a newspaper' etc., and the computer would determine the document type and apply the desired style.

You mentioned adaptive layout, and the trouble that designers have had 'getting the idea'. I agree, and I'm thinking its because adaptive layout probably isn't a design 'problem' in any traditional sense -- the sense of determining how a thing should look --, but is really an algorithmic problem: how to best fit this content into this area, and that seems to me mostly automatable if one knows what kind of content it is and how readers prefer to engage with such content.

Kevin Larson's picture

> This would involve making variants of a typeface and testing which reads best.
This paper by Professor Beier does exactly that: Design Improvements for Frequently Misrecognized Letters

hrant's picture

I think the main way onscreen reading can greatly surpass
print reading is to vary, in real-time, the size of the text
on the line being read in proportion to its distance from the
fovea; the increasingly larger boumas would overcome the
increasing blurriness, allowing for much longer saccades,
or perhaps even a mechanism where entire lines are read at
once, being rendered one at a time. The reader would no
longer even saccade; but there needs to be a way to trigger
the next line.

Kevin: that's legibility, not readability.

hhp

k.l.'s picture

John – and I meant in the visual sense of text not the literary

Text(ure) production in a quite literal sense – then we are on the same page, including the entire second paragraph.

As to the first paragraph: Given that "to make reading on screen a better experience than reading from a book" (from the text in question) is so boldly put, I would not have thought of something as simple as color. It was just normal with manuscripts and early printed pieces, and it had its come-back with website design (for visualizing the structure of a text and for indicating functionality). In so far I would turn the last sentence upside down: Not asking the theoretical question how else color might be used, but going by the assumption that each document will provoke the practical question how to visualize it.

I don't see why adaptive layout should not be a design problem. It, too, is about getting relations/proportions right, among other things. The algorithm is the implementation of the answer to the question how a thing should look. The only difference is that it is not about how an individual fixed thing should look but about how a thing that adapts to certain medium-determined conditions should look. The target is not a static but a moving one. And your "if one knows what kind of content it is and how readers prefer to engage with such content" is nothing else but "design problem".

Richard Fink's picture

@kl

Yes, the problem is conceptual. A lack of mental tools. You got it.

John Hudson's picture

Karsten: The only difference is that it is not about how an individual fixed thing should look but about how a thing that adapts to certain medium-determined conditions should look.

This prompts two thoughts.

The first is that adaptive layout doesn't need to presume certain medium-determined conditions; it can also presume uncertain conditions, and this is where I see a difference between adaptive design and adaptive process. Adaptive design tends to extend concepts from static design into an adaptive framework bounded by certain conditions. An adaptive process is one that is scalable to unknown conditions.

The second thought is that adaptability, in terms of display of electronic documents, may be a strong constraint within at least some important conditions. Now, good design thrives on constraints, or at least responds to them effectively, but strong constraints are determinative of form, and hence at once limit design options and offer themselves to automated solutions. This thought can be expressed in a question: 'How many viable options are there to lay out a particular document type adaptively on my phone?' My sense is that while there remain many design decisions to be made in the styling of documents, there are relatively few options in terms of effective layout at the smaller end of the media spectrum, i.e. where people are reading more and more. So that strikes me as something that is at once not a significant arena for design, but conversely a great area for automation.

Syndicate content Syndicate content