7 ms·

It's admirable that MS is responding to criticism of their device and working for a solution. However, all the issues he experienced seem like they would be qu

It's admirable that MS is responding to criticism of their device and working for a solution. However, all the issues he experienced seem like they would be quite obvious to anyone testing the device usage for any kind of drawing application (which, for a device with a stylus seems like a common enough use case to be testing), perhaps Microsoft needs to spend more time and effort on their QA process for the next round of the Surface instead of playing catch up after release.

stcredzero 12y ago
all the issues he experienced seem like they would be quite obvious It's the classic QA paradox. The typical hacker world view doesn't prioritize such feel issues enough. Even in today's UX focused world, stylus specific issues aren't widely understood. Combine the dynamics of the Asch conformity experiment, and it makes perfect sense. https://www.youtube.com/watch?v=FnT2FcuZaYI https://www.youtube.com/watch?v=FnT2FcuZaYI
- icebraining 12y ago
  The typical hacker shouldn't be testing this device, then. I'm sure Microsoft has someone with the skills on their teams.
  - stcredzero 12y ago
    Yes, but did the management of the team realize the above, and did that manager make sure the team would work to produce good results in this area?
  - MaulingMonkey 12y ago
    > The typical hacker shouldn't be testing this device, then. Like, say, Gabe? :)
devindotcom 12y ago
Yeah, I'm baffled as to how the button placement made it past any kind of initial hands-on testing. It's hard to imagine a person using the pen on the screen for ten or fifteen minutes not encountering that problem.
- Mister_Snuggles 12y ago
  From the article, it sounds like the Microsoft folks were kicking themselves for that oversight. To their credit, I've made similar mistakes in my career.
  - TwistedWeasel 12y ago
    Yes, me too. However i'd hope that for a product as big and important to MS as this then there would be more checks and balances in place to make sure that such things don't get overlooked. It's understandable for one engineer to overlook it, maybe even a whole team but for an entire division of design, engineering, QA, marketing etc? Something is rotten in their process.
    - lhc- 12y ago
      People make mistakes. Every piece of software ever written has bugs: I don't think that means everyone's process is rotten. It just means they are human.
      - TwistedWeasel 12y ago
        Ok, but large corporations are supposed to be able to avoid human centric mistakes with good operations workflows that provide checks and balances in their pipeline. I would assume MS is not shipping products directly from the engineering lab to the factory, so for something glaring to get all the way to the customer then there is something wrong in their process that failed to correct for human error.
        stcredzero 12y ago
        Ok, but large corporations are supposed to be able to avoid human centric mistakes They are not supposed to be worse?
    - josephlord 12y ago
      The developers were probably using non-final hardware and/or sharing engineering samples if they had samples at all rather than developing drivers to the datasheet. By the time the marketing team got their hands on them the advertising may be booked and pre-orders received from retailers (if they ever actually get their hands on products rather than just arranging final mock ups or production trial run results shipment for photo-shoots etc.). QA should have had a short window to study this type of issue but is it something that they would delay shipment over? Basically I would expect the product to ship on schedule unless there was a really critical problem and that there wouldn't be slack in the schedule for weeks of refinement. Component orders may be place 6 months ahead to secure supply so it is hard to flex the schedule without causing inventory problems not to mention messing customers about. Based on my experience in a CE company that wasn't Microsoft.
phaus 12y ago
>It's admirable that MS is responding to criticism of their device and working for a solution. It is admirable, but their work towards a solution will be fruitless. N-Trig's pressure sensitive stylus technology is vastly inferior to Wacom. Sure, most people can't tell the difference, but most artists can. Sure, artists can still produce professional quality work with an N-Trig stylus, but the experience of using an N-Trig stylus is substandard. Microsoft improved nearly every aspect of the Surface Pro 3, but their switch to N-Trig was extraordinarily stupid. I'd rather have a slightly thicker, slightly more expensive device than one with a less than perfect stylus. I hope they continue to improve things by switching back to Wacom in the future.
- wlesieutre 12y ago
  On the other hand, the N-trig stylus works accurately at the edge of the screen. You can't say the same about Wacom's, and the issue is made worse by the fact that most desktop software puts toolbars full of tiny mouse targets around the edge of the screen. With their Bamboo/Intuos/Cintiq products, Wacom can avoid the accuracy falloff problems by including an enormous margin around the edge of the active area. On a portable device like the Surface Pro 2, they weren't able to do that and it showed. There are certainly tradeoffs in going to N-trig, but I don't think it's fair to portray it as inferior to Wacom's tech in every area but cost.
- cc_ 12y ago
  The SP2 Wacom setup was not perfect by any stretch of the imagination, in my experience -- accuracy was terrible in the corners / near the edges, and parallax was a serious issue, if you did not write/draw with the pen perpendicular to the surface of the screen, there was a significant offset from the tip of the pen to where the line was actually drawn. These were bad enough to turn me (and at least one professional artist that I know) off from buying one. The move to N-Trig hypothetically fixes both of these. Early reviews/videos say that corner accuracy is greatly improved, and the lack of a separate digitizer layer allows a thinner optical stack, reducing parallax (and allowing the device overall to be thinner). Driver support has historically been an issue, but msft seems to be improving things significantly. There are fewer levels of sensitivity and hovering doesn't work quite as well, but I am overall reasonably optimistic about the switch.
omikun2 12y ago
I want to applaud MS but this whole thing stinks of a PR move. Wacom has always had a customizable pressure curve in their software. This should be a bedrock standard but instead this catch-up is spun as MS' dedication to artists. To think they are just getting around to doing this (not even a user adjustable slider scale but just a few presets) speaks volume of how much they actually care about artists -- it wasn't goal they had in mind until they sent out their demo units. Personally, I think the SP3 is probably the best PC for digital artists, but I'm disappointed in how little they worked on making the stylus user friendly.
- steele 12y ago
  Would you have preferred they did not address the concerns of Krahulik? SP2 is likely the best portable computer for digital illustrators. You can still have an SP2, but if you're like me, you're bummed out that SP3 means that the accessory ecosystem is stuck as-is for the SP2.
krschultz 12y ago
Seemingly pedantic point, but this is a UX design problem not a QA problem. QA is largely something that you want to automate, minimize, and reduce the cost of. QA is making sure things work as designed. UX design is something that you want to maximize. That's figuring out what you want to make. The tools of the UX design trade are very different. I think the best UX is done with mockups that go into customer feedback sessions. The middle ground is lots of static drawings of UX, which leads to a lot of bikeshedding and not a lot of real data. Obviously the worst is no UX design at all.
- TwistedWeasel 12y ago
  I somewhat disagree, it depends on how QA fits into your organization and how you scope its role. UX needs Quality Assurance too, and sometimes things get past the UX team and into a product - at that point new issues come to light during testing and should be fed back up to the responsible teams. In general I view QA as the last line of defense before the customer, if your QA doesn't speak up about ANY issues with the product, technical, UX or otherwise then who will? Of course, my view of the role of QA in product design may be different from others.
  - mark-r 12y ago
    The best QA people I've worked with have all had this view.
- bcbrown 12y ago
  >QA is making sure things work as designed. That is why I don't like describing the Test discipline as QA. It's not QA. When I worked in Test, I viewed my role as making sure things work well for the user, which includes 'testing' the specs to make sure there's no scenarios with glaring pain points. I think there's some parallels with the Obmudsman role in acting as the internal customer advocate.
chillingeffect 12y ago
> It's admirable that MS is responding to criticism of their device and working for a solution. It is, but it isn't some trade secret, they could open source this and have 1,000s of people converging toward the solution. Why are they so hung up and being The Ones Who Deliver The Software And The Hardware?
- virtue3 12y ago
  money? "optimal experience" I'm all for supporting open source. But seriously, it took me about 20-30 hours to get my FreeNAS setup working (the way I wanted) (getting the hardware together, configuring... lots and lots of configuring). And I'm definitely not going to say that that project is "end user friendly". I really really love open source projects but I'm willing to put down that they don't always meet the end user in an appreciable way without a serious and committed company behind the product making it end user applicable (android, et all). Otherwise the open source project will fit the needs of the people that work on it (engineers) and that's about it.
  - icebraining 12y ago
    Pretty sure chillingeffect's idea is to have Microsoft behind it, so I'm not sure how that applies.
    - virtue3 12y ago
      Sorry, I didn't really understand the parents as that. I guess a more on topic reason might be: They can barely manage a team of their own engineers to get this product out. Throwing more guys at it isn't going to help necessarily. Not without significantly more overhead and more managers. I think mystical man month definitely still applies to open source projects as well. that and microsoft is just not good at managing open source. And I can't imagine anyone would want to work on the windows kernel without being paid to do so >_>
- mpyne 12y ago
  > Why are they so hung up and being The Ones Who Deliver The Software And The Hardware? For the same reason that Ubuntu and Firefox are?

stcredzero 12y ago

all the issues he experienced seem like they would be quite obvious It's the classic QA paradox. The typical hacker world view doesn't prioritize such feel issues enough. Even in today's UX focused world, stylus specific issues aren't widely understood. Combine the dynamics of the Asch conformity experiment, and it makes perfect sense. https://www.youtube.com/watch?v=FnT2FcuZaYI https://www.youtube.com/watch?v=FnT2FcuZaYI

icebraining 12y ago
The typical hacker shouldn't be testing this device, then. I'm sure Microsoft has someone with the skills on their teams.
- stcredzero 12y ago
  Yes, but did the management of the team realize the above, and did that manager make sure the team would work to produce good results in this area?
- MaulingMonkey 12y ago
  > The typical hacker shouldn't be testing this device, then. Like, say, Gabe? :)

devindotcom 12y ago

Yeah, I'm baffled as to how the button placement made it past any kind of initial hands-on testing. It's hard to imagine a person using the pen on the screen for ten or fifteen minutes not encountering that problem.

Mister_Snuggles 12y ago
From the article, it sounds like the Microsoft folks were kicking themselves for that oversight. To their credit, I've made similar mistakes in my career.
- TwistedWeasel 12y ago
  Yes, me too. However i'd hope that for a product as big and important to MS as this then there would be more checks and balances in place to make sure that such things don't get overlooked. It's understandable for one engineer to overlook it, maybe even a whole team but for an entire division of design, engineering, QA, marketing etc? Something is rotten in their process.
  - lhc- 12y ago
    People make mistakes. Every piece of software ever written has bugs: I don't think that means everyone's process is rotten. It just means they are human.
    - TwistedWeasel 12y ago
      Ok, but large corporations are supposed to be able to avoid human centric mistakes with good operations workflows that provide checks and balances in their pipeline. I would assume MS is not shipping products directly from the engineering lab to the factory, so for something glaring to get all the way to the customer then there is something wrong in their process that failed to correct for human error.
      - stcredzero 12y ago
        Ok, but large corporations are supposed to be able to avoid human centric mistakes They are not supposed to be worse?
  - josephlord 12y ago
    The developers were probably using non-final hardware and/or sharing engineering samples if they had samples at all rather than developing drivers to the datasheet. By the time the marketing team got their hands on them the advertising may be booked and pre-orders received from retailers (if they ever actually get their hands on products rather than just arranging final mock ups or production trial run results shipment for photo-shoots etc.). QA should have had a short window to study this type of issue but is it something that they would delay shipment over? Basically I would expect the product to ship on schedule unless there was a really critical problem and that there wouldn't be slack in the schedule for weeks of refinement. Component orders may be place 6 months ahead to secure supply so it is hard to flex the schedule without causing inventory problems not to mention messing customers about. Based on my experience in a CE company that wasn't Microsoft.

phaus 12y ago

>It's admirable that MS is responding to criticism of their device and working for a solution. It is admirable, but their work towards a solution will be fruitless. N-Trig's pressure sensitive stylus technology is vastly inferior to Wacom. Sure, most people can't tell the difference, but most artists can. Sure, artists can still produce professional quality work with an N-Trig stylus, but the experience of using an N-Trig stylus is substandard. Microsoft improved nearly every aspect of the Surface Pro 3, but their switch to N-Trig was extraordinarily stupid. I'd rather have a slightly thicker, slightly more expensive device than one with a less than perfect stylus. I hope they continue to improve things by switching back to Wacom in the future.

wlesieutre 12y ago
On the other hand, the N-trig stylus works accurately at the edge of the screen. You can't say the same about Wacom's, and the issue is made worse by the fact that most desktop software puts toolbars full of tiny mouse targets around the edge of the screen. With their Bamboo/Intuos/Cintiq products, Wacom can avoid the accuracy falloff problems by including an enormous margin around the edge of the active area. On a portable device like the Surface Pro 2, they weren't able to do that and it showed. There are certainly tradeoffs in going to N-trig, but I don't think it's fair to portray it as inferior to Wacom's tech in every area but cost.
cc_ 12y ago
The SP2 Wacom setup was not perfect by any stretch of the imagination, in my experience -- accuracy was terrible in the corners / near the edges, and parallax was a serious issue, if you did not write/draw with the pen perpendicular to the surface of the screen, there was a significant offset from the tip of the pen to where the line was actually drawn. These were bad enough to turn me (and at least one professional artist that I know) off from buying one. The move to N-Trig hypothetically fixes both of these. Early reviews/videos say that corner accuracy is greatly improved, and the lack of a separate digitizer layer allows a thinner optical stack, reducing parallax (and allowing the device overall to be thinner). Driver support has historically been an issue, but msft seems to be improving things significantly. There are fewer levels of sensitivity and hovering doesn't work quite as well, but I am overall reasonably optimistic about the switch.

omikun2 12y ago

I want to applaud MS but this whole thing stinks of a PR move. Wacom has always had a customizable pressure curve in their software. This should be a bedrock standard but instead this catch-up is spun as MS' dedication to artists. To think they are just getting around to doing this (not even a user adjustable slider scale but just a few presets) speaks volume of how much they actually care about artists -- it wasn't goal they had in mind until they sent out their demo units. Personally, I think the SP3 is probably the best PC for digital artists, but I'm disappointed in how little they worked on making the stylus user friendly.

steele 12y ago
Would you have preferred they did not address the concerns of Krahulik? SP2 is likely the best portable computer for digital illustrators. You can still have an SP2, but if you're like me, you're bummed out that SP3 means that the accessory ecosystem is stuck as-is for the SP2.

krschultz 12y ago

Seemingly pedantic point, but this is a UX design problem not a QA problem. QA is largely something that you want to automate, minimize, and reduce the cost of. QA is making sure things work as designed. UX design is something that you want to maximize. That's figuring out what you want to make. The tools of the UX design trade are very different. I think the best UX is done with mockups that go into customer feedback sessions. The middle ground is lots of static drawings of UX, which leads to a lot of bikeshedding and not a lot of real data. Obviously the worst is no UX design at all.

TwistedWeasel 12y ago
I somewhat disagree, it depends on how QA fits into your organization and how you scope its role. UX needs Quality Assurance too, and sometimes things get past the UX team and into a product - at that point new issues come to light during testing and should be fed back up to the responsible teams. In general I view QA as the last line of defense before the customer, if your QA doesn't speak up about ANY issues with the product, technical, UX or otherwise then who will? Of course, my view of the role of QA in product design may be different from others.
- mark-r 12y ago
  The best QA people I've worked with have all had this view.
bcbrown 12y ago
>QA is making sure things work as designed. That is why I don't like describing the Test discipline as QA. It's not QA. When I worked in Test, I viewed my role as making sure things work well for the user, which includes 'testing' the specs to make sure there's no scenarios with glaring pain points. I think there's some parallels with the Obmudsman role in acting as the internal customer advocate.

chillingeffect 12y ago

> It's admirable that MS is responding to criticism of their device and working for a solution. It is, but it isn't some trade secret, they could open source this and have 1,000s of people converging toward the solution. Why are they so hung up and being The Ones Who Deliver The Software And The Hardware?

virtue3 12y ago
money? "optimal experience" I'm all for supporting open source. But seriously, it took me about 20-30 hours to get my FreeNAS setup working (the way I wanted) (getting the hardware together, configuring... lots and lots of configuring). And I'm definitely not going to say that that project is "end user friendly". I really really love open source projects but I'm willing to put down that they don't always meet the end user in an appreciable way without a serious and committed company behind the product making it end user applicable (android, et all). Otherwise the open source project will fit the needs of the people that work on it (engineers) and that's about it.
- icebraining 12y ago
  Pretty sure chillingeffect's idea is to have Microsoft behind it, so I'm not sure how that applies.
  - virtue3 12y ago
    Sorry, I didn't really understand the parents as that. I guess a more on topic reason might be: They can barely manage a team of their own engineers to get this product out. Throwing more guys at it isn't going to help necessarily. Not without significantly more overhead and more managers. I think mystical man month definitely still applies to open source projects as well. that and microsoft is just not good at managing open source. And I can't imagine anyone would want to work on the windows kernel without being paid to do so >_>
mpyne 12y ago
> Why are they so hung up and being The Ones Who Deliver The Software And The Hardware? For the same reason that Ubuntu and Firefox are?