Results 1 to 11 of 11
Like Tree6Likes
  • 2 Post By Preemptive
  • 1 Post By TJs11thPre
  • 2 Post By briest
  • 1 Post By Preemptive
  1.    #1  
    I thought I'd post a link to this article as it seems interesting.
    Samsung is done with Android Wear watches, says Tizen is the future | Ars Technica UK

    For technical types, there's a link in the comments to another forum:
    https://what.thedailywtf.com/topic/15001/enlightened

    It's generally assumed that the two big Korean manufacturers have webOS & Tizen as their respective insurance policies against a 'Google problem': That Google / Android will either reach the kind of dominance that will enable Google to dictate terms to them or that it somehow fails. The emergence of new 'smart' technology sectors in the wake of mobile phones has created an opportunity for these former hardware specialising manufacturers to assert greater control in the now important realm of software, operating systems and user platforms.

    LG have their TVs and experimented with a webOS watch. Samsung followed with Tizen TVs and now announce all future watches will run Tizen.

    webOS fans know that it's lack of adoption was in part due to the small app inventory - a vicious spiral that other systems have suffered from. Small user base -> lack of developer interest -> lack of apps -> Small user base.

    It seems to me that a boost for the iphone was the itunes service that preceded it - apps came later. For Android, it was a software giant offering a free and (then) open system to run on competing smartphones, saving OEM development costs while getting Google services into the hands of more users.

    Now the market has consolidated, Apple is making more money, while Google has the largest user-base. Samsung dominates the hardware business, but they and their competitors struggle to get good profit margins.

    Logic would suggest that there may be better chances in new sectors and in an increasingly connected world, products in these new markets could run whatever system as long as it can interact with others. It may even be possible to reclaim territory in mobile phones by first building an eco-system in other markets.

    I suggest that this move by Samsung is a good sign that they are trying this. If your TV, watch (and even house?) run Tizen, why not have a Tizen phone? Could Samsung become another Apple (hardware AND software)? LG and Samsung are notoriously competitive with each other. Reading the second link, I note that Tizen's system is compared with Qt (as used on LuneOS) and found wanting. LG arguably have a system better suited to connectivity and the much heralded web apps. If not mis-managed(!), LG might feel obliged to follow Samsung's moves with their own OS and find their system is more attractive to developers and users.

    So, my guess is that:
    • Using your system in new markets (building the eco-system in new territory).
    • Connecting with other systems / devices.
    • Being developer friendly.
    • Being web app friendly.

    ...is a reasonable strategy.

    For those on this forum, I'd point out that LuneOS arguably already meets the two final points. It will be able to interact with LG TVs and is already running on the Raspberry Pi (often used for home automation projects). I'd also point to Metaview's app that enables webOS phones to connect to Pebble watches. That first point though, will almost certainly have to be delivered by LG. If LG don't, LuneOS could still gain some traction thanks to connectivity and cross-platform / web apps. Of course, if LG deploys webOS beyond TVs and creates a broader eco-system, it will have also created a more favourable environment for it's own webOS phone.
    hfGermany and KURT B like this.
  2. #2  
    It would be very interesting to see the evolution of IoT brings forth a new webOS phone some day.

    Like 5-7 years from now.

    But it would probably still follow the slab form factor so IDGAF. I don't want a 5" piece of glass in my pocket.

    And i can't envision a future of people talking into their wrist to send a text message. So watches are going nowhere as a communication device. The input method is the sticking point.

    I don't think they know where to go at this point.
    Sporting my 13th Pre device, a NOS unlocked ROW Pre3!
    scjjtt likes this.
  3. briest's Avatar
    Posts
    133 Posts
    Global Posts
    134 Global Posts
    #3  
    The old Graffiti would be interesting as an input method for small and limited devices. While fingers are extremely clumsy as a pointing tool, simple glyphs should be recognizable enough...
    T5, Clié NZ90, Treo 650, Centro, Pre, Pre+, Pre 2, Pre 2, Pre2, Nexus7@LuneOS
    Preemptive and scjjtt like this.
  4.    #4  
    Quote Originally Posted by briest View Post
    The old Graffiti would be interesting as an input method for small and limited devices. While fingers are extremely clumsy as a pointing tool, simple glyphs should be recognizable enough...
    It's an interesting idea - in the right mode, the whole face could be used for finger input of the glyphs. Presumably, Graffiti is IP of the former Palm. I don't know if it is free to use or who would own the rights now... (Qualcomm is my guess).

    I note the article now has an update in which Samsung appears to deny the headline. Regardless of their policy, it would certainly be foolish to announce 'no more' Android phones if the Tizen effort fails and Samsung have to go back and eat humble pie.

    It's interesting that MS has just exited the smartphone market. They say they will still license Win10 to OEMs, but that's what was said about webOS. Blackberry seems to be in terminal decline at least in the consumer sector. What players have we now in mobile phones? iOS, Android, Cyanogenmod, the Chinese Android variant. Is BB10 likely to appear on any new models? Then there's Sailfish, Tizen and... LuneOS! Have I missed any actively developed, available SmartphoneOS?

    I Wonder if MS's developers will join Jolla? They seem to have their own problems.

    As long as LuneOS is developed at the current rate by volunteers, it may well be taken up by former webOS fans, so will have a few hundred, perhaps even a few thousand users. But any commercial project will need at minimum interoperability, connectivity or an alternative eco-system to build market share and therefore investment.

    I wrote above what I thought gave the current dominant systems their advantages. Though there will obviously be a webOS bias here, it is worth noting that BB10 and winphone both had a good reception, while Android was generally considered to be the poorest OS (for some years). Yet Palm made desperate attempts to connect to itunes and others have made efforts to run Android apps. So it would seem that quality of the UX is not what brings market penetration, but some other factor, be it apps, services or low development costs for OEMs.

    It's possible that there could be a new development in smartphones, but anything that can be replicated will be crushed by Apple or Google if not simply bought out. The winning strategy is to find a 'killer app' that can be developed and 'owned'. It will be easier to do this in new markets. Whether it's watches, cars, home automation or 'other', someone could develop something that suddenly takes off. Many IoT applications will likely require a remote interface - a phone or tablet as 'remote control'. If LuneOS retains it's DNA as an open, developer friendly, easy to use OS, then there maybe some opportunity here.

    If 'success' for LuneOS is that it is at least actively developed and capable of accessing popular, available services, then it needs a user-base large enough that there is a significant percentage of app & OS developers. I'd suggest a small virtuous circle could be built if app and OS developers consider what they can connect to and interact with. That might create an eco-system conducive to growth and the emergence of a 'killer app'.
    scjjtt likes this.
  5. briest's Avatar
    Posts
    133 Posts
    Global Posts
    134 Global Posts
    #5  
    Quote Originally Posted by Preemptive View Post
    It's an interesting idea - in the right mode, the whole face could be used for finger input of the glyphs. Presumably, Graffiti is IP of the former Palm. I don't know if it is free to use or who would own the rights now... (Qualcomm is my guess).
    No, Access Co. has bought it with PalmSource acquisition back in '00s. The bitter irony is one can now install Graffiti on Android; it's freeware but by no means open. The question is, _what_ do these IP rights cover: stroke recognition engine (most probably), glyphs shapes (possible) or the very concept of single-stroke glyphs (unlikely, if I understand correctly outcome of infamous Palm-Xerox lawsuit).

    Quote Originally Posted by Preemptive View Post
    Then there's Sailfish, Tizen and... LuneOS! Have I missed any actively developed, available SmartphoneOS?
    Ubuntu? In terms of financial/organizational backing it comes second here. Of course, Lumia demise shows it does not matter that much...
    T5, Clié NZ90, Treo 650, Centro, Pre, Pre+, Pre 2, Pre 2, Pre2, Nexus7@LuneOS
  6.    #6  
    Quote Originally Posted by briest View Post
    No, Access Co. has bought it with PalmSource acquisition back in '00s. The bitter irony is one can now install Graffiti on Android; it's freeware but by no means open. The question is, _what_ do these IP rights cover: stroke recognition engine (most probably), glyphs shapes (possible) or the very concept of single-stroke glyphs (unlikely, if I understand correctly outcome of infamous Palm-Xerox lawsuit).
    I don't think I recall this lawsuit - do tell!

    While we're here, someone tweeted this today:
    35 Interface Innovations that Rocked Our World - Xerox

    No mention of grafitti...
  7. briest's Avatar
    Posts
    133 Posts
    Global Posts
    134 Global Posts
    #7  
    Disclaimer: someone here is surely more familiar to the story, so please correct my flaky memory and possible misunderstandings.

    For what I remember, and quick "ducking" (I avoid Google) reveals, back in '90s Xerox sued 3com (let's call it Palm hereafter, together with other companies of Palm project heirloom, I really don't remember their creating, splitting and so on) over infringement of X's handwriting recognition system, called Unistroke. Graffiti was created later, had some similarities in glyph shapes, was based on the same one-stroke principle and moreover, one of Palm management was reportedly present to Unistroke demostration; Xerox won in the first instance. Palm subsequently switched to (third-party) Jot system, here called Graffiti 2 -- similar, but no longer single-stroke.
    In my opinion, Jot (G2), while more similar to natural handwriting, was also more error-prone and a bit slower to type than original Graffiti, at least on post-Xerox Palms and Symbian UIQ phones, where I had the opportunity to use it extensively. General Palm users community seemed to share this sentiment. Meanwhile the legal dance was still going on behind the scenes.
    Some years later, in mid-00s, court has ruled in Palm favour and moreover invalidated Xerox' patent, citing prior art from early '80s.
    Access got Graffiti along with PalmSource; it was going to be incorporated in next-gen PalmOS (Cobalt), based on ALP (Access Linux Platform). Cobalt was never released to the market; ALP was published on some open license, but without Graffiti.
    For what I know, Xerox never used Unistroke for anything other than harrasing Palm - not for use in their own products, not for sublicensing, not for sale.
    Last edited by briest; 06/04/2016 at 12:25 PM. Reason: of course I've mistaken some of the Palm-flock companies
    T5, Clié NZ90, Treo 650, Centro, Pre, Pre+, Pre 2, Pre 2, Pre2, Nexus7@LuneOS
  8. #8  
    Quote Originally Posted by briest View Post
    The question is, _what_ do these IP rights cover: stroke recognition engine (most probably), glyphs shapes (possible) or the very concept of single-stroke glyphs (unlikely, if I understand correctly outcome of infamous Palm-Xerox lawsuit).
    From what I know and remember, it specifically concerned the strokes of certain letters. The stroke recognition engine could not be covered simply because Palm would have implemented its own. They did not have access to Xerox' engine. The concept of single-stroke gestures (or glyphs) was well known 'long' before Xerox' patent, so that too would not have been covered. And the 'only' thing Palm changed was some of the glyphs/gestures (and they added some multi-stroke recognition, which is not trivial, but probably has nothing to do with the law suit but simply was a case of: if we have to change things, then why not develop it further?)

    One of the great things about single gesture recognition (or multi-finger single gesture recognition) is that the computational weight behind the recognition itself is quite light and fast, and will therefore work even on 'simpler' devices. For those interested in the process, and not afraid of some formulas, I would advice to take a look at Dean Rubine's paper on Specifying Gestures by Example published in 1991 (so prior to Xerox' patent), which shows both recognition and customization of gestures. There are plenty of other approaches you could use, and the truth is: as long as you're working on single strokes, things should be relatively straight-forward.

    Also, I somehow think it's a funny, that in 1991 terms like gestures and multi-finger were apparently already commonplace in the field, even though many people seem to think these are relatively recent terms/inventions (not uncommonly attributed to Apple).
  9. briest's Avatar
    Posts
    133 Posts
    Global Posts
    134 Global Posts
    #9  
    Quote Originally Posted by Misj' View Post
    From what I know and remember, it specifically concerned the strokes of certain letters. The stroke recognition engine could not be covered simply because Palm would have implemented its own. They did not have access to Xerox' engine. The concept of single-stroke gestures (or glyphs) was well known 'long' before Xerox' patent, so that too would not have been covered. And the 'only' thing Palm changed was some of the glyphs/gestures (and they added some multi-stroke recognition, which is not trivial, but probably has nothing to do with the law suit but simply was a case of: if we have to change things, then why not develop it further?)
    I wrote about current state: what does current, posessed by Access "IP rights" cover; Xerox was only mentioned as a reason why I think single-stroke principle probably is not covered (this is how I remember it, possibly wrong: that shapes were different from the start -- see https://patentimages.storage.googlea...S5596656-1.png for Xerox glyphs -- and single-stroke was the culprit).

    Some legal documents related to the case (rulings, Xerox patent application...) are available on the Net, but I'm afraid I simply cannot understand legalese well enough; it's hard in my own language and close to impossible in foreign.

    Quote Originally Posted by Misj' View Post
    One of the great things about single gesture recognition (or multi-finger single gesture recognition) is that the computational weight behind the recognition itself is quite light and fast, and will therefore work even on 'simpler' devices.
    Ah, I took Palm V from the drawer to play with for a moment -- 16MHz processor, 2MB RAM and so unbelievably snappy. Sure. today we want more from our devices, but WHY even simple tasks are slower?

    Quote Originally Posted by Misj' View Post
    Also, I somehow think it's a funny, that in 1991 terms like gestures and multi-finger were apparently already commonplace in the field, even though many people seem to think these are relatively recent terms/inventions (not uncommonly attributed to Apple).
    To be honest, Apple Newton came to handwriting recognition before Palm. With hilarious results, but still
    T5, Clié NZ90, Treo 650, Centro, Pre, Pre+, Pre 2, Pre 2, Pre2, Nexus7@LuneOS
  10. #10  
    Quote Originally Posted by briest View Post
    I wrote about current state: what does current, posessed by Access "IP rights" cover
    That is also one of the reasons why I referred to a paper from 1991. It is well before the patents related to graffiti. Since you (officially) can't patent 'prior art', the basic concept of single-gesture recognition can not be part of the IP.

    The recognition engine can, and there may be some proprietary ideas in there, but using commonly available statistics to achieve the same cannot be covered by the IP. That is also, why can be assumed that 'only' the shapes are covered.

    To be honest, Apple Newton came to handwriting recognition before Palm. With hilarious results, but still
    But...the paper I cited was from 1991 (which itself referred to earlier papers) while the Newton was from 1993. And it was the paper that used the terms gesture and multi-finger.


    Another this is, that strictly speaking Palm has gesture-recognition, not handwriting-recognition. This is part of the genius, since it solves many problems like: a much smaller predefined set to compare your gestures to (letters can be written in many different ways, gestures cannot), single stroke vs. multi-stroke (multi-stroke is much more complex because you have to determine whether a stroke is part of the previous letter or the next letter), it is much more scale independent (or at least easier to take scale into account), capitals and and lowercase can use the same gesture and can be distinguished by entering a 'capital'-mode (with handwriting the l (lowercase L) and I (uppercase i), and 1 are difficult to distinguish), etc. etc.

    In short, if someone wanted to implement a form of graffiti, neither the recognition of gestures, nor the registration of gestures are patented or anyone's IP. The specific shapes of the glyph gestures are Access' IP, so that would be a risk. On the other hand one has to wonder whether these gestures are as useful as finger-strokes as they use to be with a stylus.
  11.    #11  
    Quote Originally Posted by briest View Post
    The old Graffiti would be interesting as an input method for small and limited devices. While fingers are extremely clumsy as a pointing tool, simple glyphs should be recognizable enough...
    You called it! Now on Apple watchOS...

    I think they're calling it scribble or something similar.

Similar Threads

  1. Copy-cating all over again (Samsung Tizen)
    By GoJoe2 in forum LG webOS TV
    Replies: 4
    Last Post: 05/17/2015, 05:43 PM
  2. Will LG make a WebOS after Samsung did a Tizen`s?
    By akitayo in forum General News & Discussion
    Replies: 5
    Last Post: 02/09/2015, 12:12 AM
  3. HP Touchpad Samsung Tizen OS?
    By lightinside87 in forum HP TouchPad
    Replies: 4
    Last Post: 01/16/2015, 06:37 PM
  4. Tizen: a new WebOS
    By karstenskov in forum webOS Discussion Lounge
    Replies: 32
    Last Post: 05/12/2012, 04:50 AM

Tags for this Thread

Posting Permissions