Thursday, March 13, 2014

Try DBDOC on a Tablet: It Just Works!

Recently we brought a Surface Pro tablet into our office, and we've been pleasantly discovering that for the most part, DBDOC Just Works! Read on for more details.
Lets look at Hyperview, which is the most likely of the DBDOC programs to be used with a tablet.

First up, we have an option in Hyperview that happens to have prepared us for touchscreens entirely unintentionally. If you go into "Options..." you'll see in View the checkbox for "Large icon toolbar", which scales up the main buttons.


Who knew we were preparing for touchscreens all this time? It's nice to be forward-thinking by accident!

Most navigation works as you'd expect. Tap on the screen for left-click; long-press for a right-click, for example to call up a context menu on a block; scroll by flicking within a pane or by pressing down on a scrollbar and moving your finger. It can be a bit inconsistent sometimes, and we're investigating an issue where kinetic scrolling in the index pane stops working on some M14s, but in general it's quite fine with fingers or a stylus.

Selecting an area to zoom into can be a bit tricky with the touchscreen, although I've found it to be fairly reliable if I start first with a distinctly horizontal motion, then start moving my finger vertically as well. Otherwise, Windows sometimes interprets the motion as an attempt to drag the viewspace. With the stylus that comes with the Surface Pro, however, box selection seems to work without hassle.

If you're running Windows 8.1 on a relatively high-DPI device like the Surface Pro or a new high-end laptop, you may notice the fonts are a bit blurry. This is because Windows 8.1 by default scales up the UI element sizes when using a dense display, which is handy for keeping things readable (and for making it easier to touch interactive elements), but isn't done in a way that's terribly kind to applications built on the old standard Windows assumption of 96dpi. Right now, most third-party applications for Windows react in one of two ways: either upscaling with blurry fonts and other elements, or ignoring the scaling entirely. You can see this contrast with the current versions of Google Chrome and Mozilla Firefox for Windows, where the former looks the same as DBDOC does and the latter stays crisp but small in defiance of Windows' scaling settings.

If you haven't noticed this (in which case, sorry for pointing it out!) you can leave it as-is, but if you're looking to disable scaling globally, you can find the option in "Control Panel\Appearance and Personalization\Display". You can also disable in on a per-application basis by right-clicking on a shortcut or an EXE, going into its properties, and choosing "Disable display scaling on high DPI settings" from the Compatibility tab.


We're investigating and evaluating how we might hook DBDOC into the scaling functions of contemporary Windows, which will not only help with tablets like the Surface Pro, but also high-resolution desktop and laptop displays which are just now starting to edge into mainstream production; my own laptop has a resolution of 2560x1700 on a mere 13" screen!

Well, that's the rote details of the current state of running Hyperview on Windows 8/8.1 tablet, but what's the actual use? Well, we've imagined that it could be quite helpful carrying around a plant, looking at live data (if you're connecting in through a local network or VPN) or even just to be able to easily browse your graphics and CAD sheets while standing in front of the actual machinery they represent. You could use CIUMon Relay to pass on data from a running CIUMon instance, or even hook up a USB-to-serial adapter to your tablet and plug in directly to a CIU with your tablet. And the guts of Microsoft's Surface Pro and Surface Pro 2 are fairly hefty, so you could even run BuildPlus to generate M14s if need be.

We're just getting started on touchscreen and tablet support, so let us know what you would find useful and in what scenarios you can see using DBDOC on a tablet!


  1. Hi Geoff

    I am very pleased that you have started a discussion on this on your blog. I am all for using technology to improve visibility and usability of software, applications and documentation of all variants. I think the move you are suggesting to Hyperview on a tablet is fantastic and one as I previously discussed with you. I suppose it makes sense to focus on a Windows platform for development first. In the future other tablet platforms would be good too, such as iOS and Android. This way you can maximise the inherent advantages of these mature mobile device platforms. Windows is a good start.

    As far a uses for a tablet deployment of Hyperview I can see a number that I would use.
    1: Using Hyperview for commissioning check out functions. This assumes that live data is available. Using Hyperview to monitor the configuration and the values in engineering units displayed in the DCS or HMI would be great for this functionality check.
    2: As a maintenance tool for troubleshooting this would be really advantageous to have Hyperview on a mobile device. Verification of field functionality against the running configuration would save precious time in travelling from the field to the EWS and back several times.
    3: Installation check out is another one. With network access to many linked documents potentially available, depending on how extensively a site has linked documents etc to a build, someone could verify the physical installation of a loops wiring by referring to the drawings on the tablet. With Hyperview's annotation ability notes can be kept on each drawing highlighting completeness and or problems for resolution. Having just one small device to carry around would provide less paperwork to carry. With nightly builds the latest documents would be available every morning with continuity of annotations across various versions of drawings and builds.

    As a comment on item 3 above this is where your future plans to centrally store the annotations for universal availability would be fantastic. You could even choose a group to which the annotations would belong to so that only members of that group would see the annotations. This annotation availability means that those who follow on for the next shift may refer to the applicable annotations to continue work without having to worry about what has and hasn't been checked and what the issues were.

    The use of a mobile device for those of us with very limited wireless deployment on the factory floor means we need to use cellular connections. Therefore tablets with SIMs are a must for us. But here lies a problem. How to ensure data transfer is kept to a minimum for maximum functionality and speed. The latest M14 file must be transferred to the tablet or mobile device and Hyperview run on the mobile device.

    I think with any use of a technology like this management of it is very important. The potential uses need to be considered so that the application can be developed accordingly.

    This is a fantastic start and I hope that I can be actively involved in this development path as I have been with other developments. Keep up the good fight Geoff and team.

    Specialist Process Engineer
    CHH Tasman, Kawerau

    1. Andrew: In terms of SIMs and cellular connections for tablets, at the moment Microsoft has stated that their Surface Pro line won't receive a cellular-capable variant (they're launching a cellular variant of the Surface, but that runs Windows RT, so desktop applications like ours can't be installed). However, these days nearly every smartphone ships with the ability to act as a wireless hotspot (often as easy to set up as tapping a checkbox), the only restrictions being those imposed by your cellular provider. Setting up bluetooth tethering is similarly quite doable, although slightly more complicated and device-specific. So even with a non-cellular tablet, one should be able to piggyback on a cellular connection fairly easily.

      That being said, there *are* some manufacturers filling that niche, for example Lenovo makes 3G variants of its ThinkPad Tablet 2 and ThinkPad Helix lines. But it still seems quite common for Windows tablets to lack cellular connectivity, the assumption seeming to be that one would just tether--and full Windows, power-hungry as it is, might not last very long on a 3G/4G connection, while an extremely nearby wifi access point (ie. your cellphone in your pocket acting as a hotspot) can be much less of a power draw, and bluetooth is even better in that respect. At least, that's what I'd suppose is behind the reticence of OEMs to make tablets with both fullblown Windows _and_ a cellular radio.

      Which leads to talk of other OSes; it's worth mentioning that on Android, iOS and Windows RT there are remote desktop apps, in fact the one I use on my Android tablet is officially released by Microsoft themselves:

      Now, admittedly streaming the visuals of an entire desktop requires a non-trivial amount of data, so may not be appropriate in all circumstances, but if you find yourself wanting to use an iPad or an Android tablet or a Windows RT tablet (like the Nokia Lumia 2520) that has a cellular connection, it's a possible solution. And, any moves we make to improve the experience of DBDOC on a Windows 8.x touchscreen device should benefit this scenario too.

    2. Thanks Keith. I will definitely set aside some time to see how the RDP option works out for me. Did you use a Surface Pro or a Surface Pro 2? There is a bit of difference between the two devices including price.


  2. Actually, Keith Zubot-Gephart at our office did the post and work, so credit goes to him. Keith really strengthens us with his expertise in a full range of systems. I can usually take his sceptical attitude towards Windows and IE, too, and it keeps us thinking.

    "Windows is a good start." I concur, but assure you it is probably the end, too. Everything we have done in DBDOC is based on Window OS. Should ABB move to something else, we might want to follow. We have limited enough development and testing resources, so we cannot easily do significant work aimed at a small subset of users.

    Indeed, interaction with you helped us decide to purchase a Surface Pro, but it was another DBDOC user simply seeking justification to get a tablet that could do Hyperview that really triggered things. I told him to try it and see how it works, which I realized did not fit our normal development, testing and support paradigm, so we bit the bullet and bought one.

    The motivation for tablet deployment is trivial extension of DBDOC to add the tablet capability to what our clients are using right now - desktops, Citrix based interfaces and laptops (notebooks, too).

  3. Thanks for the comments, Andrew.

    We are moving slowly but surely towards several different means of supporting the sharing Hyperview user data (annotations, bookmarks, watch window groups, and now, error browser state -- stars, checks etc) among users.

    The ability to one-off import annotations and bookmarks belonging to another user has already been implemented for the next minor release 10.5.1

    The ability to share these types of data in an ongoing manner via a central repository is currently under active development and is planned for 10.5.1 as well.

    Storing collected trend data centrally will be the next thing in this area to be attacked. Unlike the small data files in the first category, the amounts of collected data can be very large, so a thoughtful strategy for shared watch window trend data is required.

  4. Wow. Thanks for your replies. I only just came back in to take a look and see if there were any updates to the blog and suddenly I find all your comments that I didn't know were there. Maybe I should have ticked the Notify me box.

    I am going to extend this a bit further and suggest that maybe Hyperview could become suitable for a web browser environment. That way if you use something such as HTML5 the device interoperability should become trivial. Maybe this is not something you could even tackle as a small company or even want to consider. However being web based has other advantages if mobile support is something you value highly for your current and future users.


    1. We've long talked and debated internally about Hyperview becoming a web/HTML-based frontend; as you surmised, it's a potentially large undertaking, so there are no current plans for it, but it's certainly a tantalizing possibility (in fact, although it's before my time at GMCL, I've heard that it was dreamed about before HTML5 even existed!).