5 ms·

These comments go hand in hand with another observation he made in a talk of his (or several), how the first books that were printed on the printing press resem

These comments go hand in hand with another observation he made in a talk of his (or several), how the first books that were printed on the printing press resembled the books before the printing press. It took people decades to figure out they can print smaller, easily portable books because they did not have to be viewed as these precious hand made things. In the same way, the web is an imitation of a paper document. There are some improvements, you can now e.g. read the New York Times from anywhere on the globe, but overall it could do so much more than that.

Kinrany 4y ago
A lot of the design constraints remain the same. For example, all three of hand-crafted books, printed paper and web pages need a specific optimal line width.
- chriswarbo 4y ago
  Browsers wrap lines automatically, so line width can be chosen by the reader, and even varied depending on the situation (the same applies to font, size, colours, etc.).
  - coldtea 4y ago
    Ideally the reader should still not be bothered and sensible defaults should be applied (even if they are dynamic defaults). Not to mention that readers left to their own devices tend to chose stupid unoptimized color schemes and font sizes all the time, hurting their readability and lowering their reading times... (of course for disabled readers it makes sense to be able to set to whatever suits them, but those also need sensible defaults for their case, perhaps even more so than abled readers, as it can be more difficult to them to change configurations)
    - TeMPOraL 4y ago
      > Not to mention that readers left to their own devices tend to chose stupid unoptimized color schemes and font sizes all the time, hurting their readability and lowering their reading times... ... thus reducing their exposure to our content marketing texts and/or ads embedded on the page, thereby impacting our bottom line! I think it's safe to assume most people capable of choosing "stupid unoptimized color schemes and font sizes" are also capable of iterating on their choice, or reverting to the better defaults, when they feel the reading experience of things they want to read worsens.
      - coldtea 4y ago
        >I think it's safe to assume most people capable of choosing "stupid unoptimized color schemes and font sizes" are also capable of iterating on their choice, or reverting to the better defaults, when they feel the reading experience of things they want to read worsens. You'd be very surprised
_glass 4y ago
Aren't Single Page Applications implementing most of the ideas? Sending programs (Javascript), and objects (JSON). I think we moved already to very different formats than print, for better or worse, i.e. endless streams vs. a limited paper resource.
- mike_hearn 4y ago
  In the 80s/90s era that these ideas are rooted in, an "object" means both code and data fused together. In this worldview JSON is not a way to send objects despite the name, it's a way to send structs. In the sense Kay means, sending an object implies receiving a thing that implements one or more interfaces, on which you call methods/send messages, and then stuff happens. The data held by the object is encapsulated. What Java was trying to do with RMI, classloaders, etc was a lot closer to this vision because in that mechanism you could actually receive over the network a full object with both code and data, in which the actual underlying data structure was something you'd never seen before but it didn't matter as long as it implemented an interface you recognized, because the code to process the data came with it. This vision hit a number of practical engineering problems, most obviously security and less obviously the pervasive assumption of flat fast LAN-like networks that the CORBA/RMI/DCOM/etc people kept making. The web won out over this vision partly because it was simple enough to reimplement. When Tim Berners-Lee didn't have the resources to take it to the full potential (not even having proper inline images) NCSA Mosaic took over, then Netscape took over from them, then Microsoft from Netscape, then Chrome/Safari from Microsoft. The history of the web is basically a history of failed business models up until Google decides to just subsidize the whole thing in order to drive search traffic.
  - pjmlp 4y ago
    Only to have anyone now use REST and gRPC for the same kind of purposes. The only major difference between what I was doing with RMI/CORBA and DCOM and now, is the tooling and the protocols. Everything else, IDL files, schemas, boilerplate generation, client SDK to wrap boilerplate code,.... is just the same story with a different dressing.
    - 4y ago
Archelaos 4y ago
> It took people decades to figure out they can print smaller, easily portable books because they did not have to be viewed as these precious hand made things. This is wrong. The first thing Gutenberg printed after the prestigious bible (which not only resembled previous bibles, but surpassed them all in respect of the regularity of the printing types), was the Donatus, Aelius Donatus' Latin Grammar book "Ars minor". This was a standard text book of the time. It was short (aprox. 30 pages) and extremly popular: There were at least 24 editions published until 1468. From 1454/1455 we have indulgence certificates, which shows that the printing press was immediately used for cheap mass production, were "precious hand made things" were of minor improtance. The next things Gutenberg printed (1455/1456) fell into the "news and politcal pamphlet" category: a 6 page long Calendar Against the Turkes ("Eyn manung der cristenheit widder die durken") and a 25 page long German translation of a papal bull also against the Turks.
- ModernMech 4y ago
  I think the parent is misremembering Kay’s point a little. It’s not that people didn’t catch on to the idea that the printing press could make books cheaper/more portable because they weren’t so laborious to create. That much was obvious about its utility. Kay’s point as I remember was that, at least initially, automation of previous tasks is all they wanted to do with it. They got this great new machine that opened up tons of possibilities, but all they used it for was to make faster the same thing they were doing before. But we all know rather than saving some monks some time, what the printing press really did was fundamentally transform society. That’s what people missed, and it’s something people keep missing about transformative technologies, even today. In his book Mindstorms (which every programmer needs to read), Seymour Papert makes the same observation about computers when they were introduced in classrooms; he says that instead of unlocking the full potential of those machines, educators just used them to automate activities they were doing before - times tables, essay writing, drawing graphs, etc. Very few people realized the potential at the time, and most still don’t.
- bazoom42 4y ago
  I love when plausible sounding factoids is dismantled by people who actually know history.
robenkleene 4y ago
I actually find the arguments to be stronger for the opposite lesson. Some examples of successful designs from the physical world that are also successful in purely digital interfaces: How smart watches display the time, VST instruments (e.g, https://mixedinkey.com/captain-plugins/wiki/best-vst-instruments-2020/ https://mixedinkey.com/captain-plugins/wiki/best-vst-instrum...), ebook "pages", on-screen keyboards, visual-coding environments that use a "cables" metaphor (e.g., https://twitter.com/robenkleene/status/1280182525013475330 https://twitter.com/robenkleene/status/1280182525013475330). I actually think the lesson here is that if you can take a metaphor from the physical world, and can incorporate it's essence into your digital design, it'll probably improve your design. And I don't think the reason is because we're leaning on our familiarity with the physical world (too many purely digital designs have proved successful for that to be the case). I think the reason is simple: Our brains are evolved for the physical world, and we've had thousands of years to master building things with physical materials, so the human race has created truly fantastic interfaces in the physical world. If we can bring the essence of those designs into the digital world, we should.
morphle 4y ago
I'll try to link to several of Alan's talks [1] referred to by localhost on the first books and printing press. This first one [2] is too short, there are more detailed versions. Talk [4] has different slides then I remember. [1] https://tinlizzie.org/IA/index.php/Talks_by_Alan_Kay https://tinlizzie.org/IA/index.php/Talks_by_Alan_Kay [2] https://vimeo.com/6606439 https://vimeo.com/6606439 [3] https://vpri.org/writings.php https://vpri.org/writings.php [4] https://tinlizzie.org/IA/index.php/Alan_Kay_at_Xerox_PARC_(1993) https://tinlizzie.org/IA/index.php/Alan_Kay_at_Xerox_PARC_(1...

Kinrany 4y ago

A lot of the design constraints remain the same. For example, all three of hand-crafted books, printed paper and web pages need a specific optimal line width.

chriswarbo 4y ago
Browsers wrap lines automatically, so line width can be chosen by the reader, and even varied depending on the situation (the same applies to font, size, colours, etc.).
- coldtea 4y ago
  Ideally the reader should still not be bothered and sensible defaults should be applied (even if they are dynamic defaults). Not to mention that readers left to their own devices tend to chose stupid unoptimized color schemes and font sizes all the time, hurting their readability and lowering their reading times... (of course for disabled readers it makes sense to be able to set to whatever suits them, but those also need sensible defaults for their case, perhaps even more so than abled readers, as it can be more difficult to them to change configurations)
  - TeMPOraL 4y ago
    > Not to mention that readers left to their own devices tend to chose stupid unoptimized color schemes and font sizes all the time, hurting their readability and lowering their reading times... ... thus reducing their exposure to our content marketing texts and/or ads embedded on the page, thereby impacting our bottom line! I think it's safe to assume most people capable of choosing "stupid unoptimized color schemes and font sizes" are also capable of iterating on their choice, or reverting to the better defaults, when they feel the reading experience of things they want to read worsens.
    - coldtea 4y ago
      >I think it's safe to assume most people capable of choosing "stupid unoptimized color schemes and font sizes" are also capable of iterating on their choice, or reverting to the better defaults, when they feel the reading experience of things they want to read worsens. You'd be very surprised

_glass 4y ago

Aren't Single Page Applications implementing most of the ideas? Sending programs (Javascript), and objects (JSON). I think we moved already to very different formats than print, for better or worse, i.e. endless streams vs. a limited paper resource.

mike_hearn 4y ago
In the 80s/90s era that these ideas are rooted in, an "object" means both code and data fused together. In this worldview JSON is not a way to send objects despite the name, it's a way to send structs. In the sense Kay means, sending an object implies receiving a thing that implements one or more interfaces, on which you call methods/send messages, and then stuff happens. The data held by the object is encapsulated. What Java was trying to do with RMI, classloaders, etc was a lot closer to this vision because in that mechanism you could actually receive over the network a full object with both code and data, in which the actual underlying data structure was something you'd never seen before but it didn't matter as long as it implemented an interface you recognized, because the code to process the data came with it. This vision hit a number of practical engineering problems, most obviously security and less obviously the pervasive assumption of flat fast LAN-like networks that the CORBA/RMI/DCOM/etc people kept making. The web won out over this vision partly because it was simple enough to reimplement. When Tim Berners-Lee didn't have the resources to take it to the full potential (not even having proper inline images) NCSA Mosaic took over, then Netscape took over from them, then Microsoft from Netscape, then Chrome/Safari from Microsoft. The history of the web is basically a history of failed business models up until Google decides to just subsidize the whole thing in order to drive search traffic.
- pjmlp 4y ago
  Only to have anyone now use REST and gRPC for the same kind of purposes. The only major difference between what I was doing with RMI/CORBA and DCOM and now, is the tooling and the protocols. Everything else, IDL files, schemas, boilerplate generation, client SDK to wrap boilerplate code,.... is just the same story with a different dressing.
  - 4y ago

Archelaos 4y ago

> It took people decades to figure out they can print smaller, easily portable books because they did not have to be viewed as these precious hand made things. This is wrong. The first thing Gutenberg printed after the prestigious bible (which not only resembled previous bibles, but surpassed them all in respect of the regularity of the printing types), was the Donatus, Aelius Donatus' Latin Grammar book "Ars minor". This was a standard text book of the time. It was short (aprox. 30 pages) and extremly popular: There were at least 24 editions published until 1468. From 1454/1455 we have indulgence certificates, which shows that the printing press was immediately used for cheap mass production, were "precious hand made things" were of minor improtance. The next things Gutenberg printed (1455/1456) fell into the "news and politcal pamphlet" category: a 6 page long Calendar Against the Turkes ("Eyn manung der cristenheit widder die durken") and a 25 page long German translation of a papal bull also against the Turks.

ModernMech 4y ago
I think the parent is misremembering Kay’s point a little. It’s not that people didn’t catch on to the idea that the printing press could make books cheaper/more portable because they weren’t so laborious to create. That much was obvious about its utility. Kay’s point as I remember was that, at least initially, automation of previous tasks is all they wanted to do with it. They got this great new machine that opened up tons of possibilities, but all they used it for was to make faster the same thing they were doing before. But we all know rather than saving some monks some time, what the printing press really did was fundamentally transform society. That’s what people missed, and it’s something people keep missing about transformative technologies, even today. In his book Mindstorms (which every programmer needs to read), Seymour Papert makes the same observation about computers when they were introduced in classrooms; he says that instead of unlocking the full potential of those machines, educators just used them to automate activities they were doing before - times tables, essay writing, drawing graphs, etc. Very few people realized the potential at the time, and most still don’t.
bazoom42 4y ago
I love when plausible sounding factoids is dismantled by people who actually know history.

robenkleene 4y ago

I actually find the arguments to be stronger for the opposite lesson. Some examples of successful designs from the physical world that are also successful in purely digital interfaces: How smart watches display the time, VST instruments (e.g, https://mixedinkey.com/captain-plugins/wiki/best-vst-instruments-2020/ https://mixedinkey.com/captain-plugins/wiki/best-vst-instrum...), ebook "pages", on-screen keyboards, visual-coding environments that use a "cables" metaphor (e.g., https://twitter.com/robenkleene/status/1280182525013475330 https://twitter.com/robenkleene/status/1280182525013475330). I actually think the lesson here is that if you can take a metaphor from the physical world, and can incorporate it's essence into your digital design, it'll probably improve your design. And I don't think the reason is because we're leaning on our familiarity with the physical world (too many purely digital designs have proved successful for that to be the case). I think the reason is simple: Our brains are evolved for the physical world, and we've had thousands of years to master building things with physical materials, so the human race has created truly fantastic interfaces in the physical world. If we can bring the essence of those designs into the digital world, we should.

morphle 4y ago

I'll try to link to several of Alan's talks [1] referred to by localhost on the first books and printing press. This first one [2] is too short, there are more detailed versions. Talk [4] has different slides then I remember. [1] https://tinlizzie.org/IA/index.php/Talks_by_Alan_Kay https://tinlizzie.org/IA/index.php/Talks_by_Alan_Kay [2] https://vimeo.com/6606439 https://vimeo.com/6606439 [3] https://vpri.org/writings.php https://vpri.org/writings.php [4] https://tinlizzie.org/IA/index.php/Alan_Kay_at_Xerox_PARC_(1993) https://tinlizzie.org/IA/index.php/Alan_Kay_at_Xerox_PARC_(1...