[TwistedWeasel]

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.

[krschultz]

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]

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]

The best QA people I've worked with have all had this view.

[bcbrown]

>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.

[stcredzero]

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

[icebraining]

The typical hacker shouldn't be testing this device, then. I'm sure Microsoft has someone with the skills on their teams.

[stcredzero]

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]

> The typical hacker shouldn't be testing this device, then.

Like, say, Gabe? :)

[phaus]

>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]

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_]

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.

[wlesieutre]

Website is having some stylesheet trouble for me, but Surface Pro Artist says they have a decent handle on the driver compatibility. The update isn't generally released yet, but it's a significant fix.

http://surfaceproartist.com/blog/2014/6/9/n-trig-closing-win...

[dman]

The possibility also remains that Wacom did not want the kind of business that the Surface brings to them. The surface is the first product that does a reasonably good job of cannibalizing cintiq sales. I would expect their margins on Cintiq to be higher than selling a digitizer component to Microsoft. I think Wacom is in a really tough place here - they have to balance between limiting access to their crown jewels while also making sure that ntrig doesnt make too many inroads as a legitimate alternative.

[BrandonLive]

Absolutely nothing to do with that. Wacom would love to be in the SP3. But weight, thickness, cooling, display quality (and writing feel due to extra layers between the glass and display panel), and battery life all conspired against them. The N-Trig solution trades off some drawing precision (at a degree very few will notice) and the requirement of a battery in the pen for improvements to all of those things. Seems like a no-brainer.

When I was at MS folks were working super closely with Atmel (touch panel vendor for most early Win8/RT tablets) and really pushed the limits of their technology. There was a tight feedback loop there and I personally had found issues which eventually were solved via iterations of back-and-forth with Atmel and with software tweaks to work around hardware limitations.

It may not happen over night, but I suspect they're doing the same thing with N-Trig and pushing them to improve the experience in a way which other PC/tablet vendors never have or would. So don't assume that just because other OEMs haven't cared enough to get the most out of N-Trig's text that Surface doesn't have a shot at doing better.

[benihana]

>It is admirable, but their work towards a solution will be fruitless.

Any Microsoft employees listening: Go ahead and pack it up. Someone on the internet has told you all you need to hear. It's fruitless. Literally nothing you do will work, as this post has clearly pointed out. Look to this non sequitor full of opinion presented as fact for all the info you need: Until you switch to the hardware that OP knows is superior, your work will be for naught. Sorry.

[phaus]

It is a fact that both Wacom and N-Trig digitizers have different sets of advantages and disadvantages. It is a fact that the Surface Pro 3 can't completely eliminate its chosen technology's disadvantages. I never claimed that the entire Surface product line was doomed to failure, after all, the world doesn't revolve around digital artists.

As a couple of people pointed out, Wacom has a major disadvantage as well. It really sucks that it loses its accuracy at the edges of the screen, but in spite of that I still find the experience far superior to using an N-Trig stylus. I have come to this conclusion after using many different devices with different kinds of digitizers over the last decade. I also frequent a few of the major online digital art communities, and they all seem to agree with me. Perhaps the N-Trig fans are just quiet, but I'm inclined to think the lack of representation is due to the fact that artists enjoy using Wacom products.

I'm not sure why you feel the need to be so childishly hostile. I wasn't aware that expressing an opinion in an anonymous online community where people gather to have casual conversations was frowned upon.

[devindotcom]

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]

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.

[stinos]

Actually the beginning of the first paragraph (repeated below) really made me think 'wtf, apart from the number of people, this is exactly how some meetings with users work at our tiny startup'. Then I realized every engineer/designer/... probably makes such mistakes, maybe becasue of losing sight on the bigger picture, and MS is no different.

I ended up in a conference room with about half a dozen people from the Surface team. More rotated in and out as I worked. I drew and talked for two and half hours while they watched and took notes. Within the first thirty seconds they realised how frustrating the home button placement was.

[TwistedWeasel]

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.

[josephlord]

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.

[lhc-]

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]

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]

Ok, but large corporations are supposed to be able to avoid human centric mistakes

They are not supposed to be worse?

[omaranto]

> To their credit, I've made similar mistakes in my career.

What did you mean? You don't mean that if X makes a mistake that you made first, that mistake is a credit to X, do you?

[Mister_Snuggles]

Not at all. Just that mistakes happen to everyone, even the biggest corporations.

I could have phrased that a little better. Perhaps "I've made similar types of mistakes that were obvious in retrospect, so I have a hard time faulting them" might be a better way to say it.

[omaranto]

Oh, OK. That makes sense.

[rickyc091]

They probably didn't have an artist use the device prior to release. They were probably testing the app in OneNote where you could just move the page to center it if you needed more room.

[TwistedWeasel]

A lot of QA departments these days put their effort behind automated testing, which is very valuable but will never replace actual human usage of a device.

Automation should catch one class of problems but real world usage is necessary to catch a whole different type of issue.

[vinceguidry]

I've worked in a testing lab for a product somewhat like the Surface. They had an automated testing platform but it was pretty much useless. They didn't put enough effort into it to make it worthwhile. As a result basic sanity testing had to be done by hand. It took forever and as a result there was no real time for 'actual human usage'.

I found the experience invaluable in helping me to understand why I found virtually every hardware and the vast majority of software products extremely painful to use. And how much I appreciate Steve Jobs for showing us how it's done.

[WalterBright]

I have that problem with every laptop with the touch pad square in front of the keyboard. I am always brushing against it with my palm while typing that produces very unintended input.

I'm not the only one, I've noticed other people having the same problem.

I have no idea how this design became ubiquitous. It does not work for me at all.

[prawn]

There's a comment from another sketcher in here talking about how it might be a peculiar grip not used by all artists.

[chillingeffect]

> 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]

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]

Pretty sure chillingeffect's idea is to have Microsoft behind it, so I'm not sure how that applies.

[virtue3]

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]

> 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?

[icebraining]

Firefox isn't delivering the hardware. And the software is open source.

[omikun2]

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]

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.