Discussion: OCAD shifts to subscription model

Am I missing something. I was intending to update to 12 but have reconsidered. I expected an annual subscription to be quite a bit less expensive than a perpetual license.

If you want to stay at the leading (bleeding?) edge you have to update OCAD every 2 years or so, I'm pretty sure (i.e. guessing) the 3-year subscription have been calculated to more or less result in the same revenue.

I know that some OCAD users have been skipping every other version since the upgrade from either 10 or 11 to 12 was similar, but now I see that the upgrade offer has a 40% discount from 12 and 20% from 11. This probably means that for an upgrade you should buy the 3-year package to maximize the rebate?

For me the new 2018 has two big advantages:

a) The check map minimum symbol distances function will make it far easier to verify that maps for major events in Norway are actually according to spec.

b) TPI-based alternative contour generation: Here it looks like I can generate both a set of raw contours for surveying as well as a quite intelligently filtered set that I can directly use as a basis for the final contours. I.e. with the raw 1m contours as a basis to decide where I need to insert form lines and/or modify the filtered contours to emphasize important features.

There has been a few Master/PhD papers from Finland that show how TPI manages to smooth out most of the insignificant noise/squiggles in flat areas while leaving the contours close to raw in steeper and more interesting terrain.

This will almost certainly get more updates in the future, I have already seen followups to the first TP papers, where they try to automatically detect where form lines will do the most good. Kartapullautin tries to do something similar.

a) and b) do sound useful, though as mentioned kartapullautin does something similar to b) for free. I thought some free versions of a) exist? Charging this much for stuff that's available free via OOM, KP, Purple Pen, various free orienteering, GIS and lidar tools, seems hard for clubs that need to manage licenses. How much do mappers find they use OCAD features that aren't available in free tools? Or do some mappers find it simply worth having an all-in-one program?

Terje, are you referring to the Kettunen / Koski / Oksanen papers?

BTW, the first one is keen orienteer, participated World Masters SkiO Championships Craftsbury in M35 and did pretty well there. So obviously they area aware of this forest navigation perspective and not just focus on nice decorative look of contours.

" I'm pretty sure (i.e. guessing) the 3-year subscription have been calculated to more or less result in the same revenue."

It must be a sophisticated pricing model that includes behavioural assumptions that lead to a decrease in user base.
Last price of OCAD12 from Australia $777 (inc GST)
New three year license from Australia $1706 (inc GST)
Averages at $569 per annum.
If we assume updates in past were every two years, then that is $1138 for the useful period of old perpetual license.
Looks to me like a 45% price rise (obviously numbers vary depending on the renewal period assumption).
I imagine the argument is that with the subscription you get updates on a regular basis. I suspect however that the change will be self-defeating. Most of my mapping is unpaid. I was finding it hard to justify the bi-annual upgrade but was ready to shift from version 10 to 12 because of compliance checking. I imagined the subscription model would operate on a similar price ratio to other perpetual to subscription conversions I experienced. I have now re-evaluated. And I already know of one other mapper who has decided to stay with 12.
The attraction of OCAD 12/2018 was the ability to use the Surface pen and the compliance checking. There are sufficient open source or freeware tools for lidar processing such that the lidar updates in OCAD do not sway my decision. For me the financial choice is to use single annual subscription fee for OCAD to instead buy an Android tablet with stylus and use OOM in the field. It would not have been my first choice with a different pricing structure. Presumably compliance checking will appear in OOM at some point. I hold less hope for surface pen compliance.

In case Terje is referring to those paper, well, back when they were developing their contour generation method they needed something to compare with. KP had some reputation for doing something, so they asked me to produce some reference contours. Makes it easier to see what could possibly be improved if you have something to compare with something made from the very same input data. So I used little time to get generalization about to the to the desired level and produced some contours. Hopefully they got some pointers from along the way, even if direct comparison aren't entirely fair really.

I'm sad to see that they have effectively raised the prices for most O users of OCAD.

As most of you probably know, my own tools for base map generation have been carefully configured to only use totally free tools, with the resulting DXF and image files directly usable by both OCAD and OOM.

With OCAD 2018 I have reprocessed a couple of recent maps, generating both raw contours every 1.0 or 0.5m as well as TPI-filtered main and index contours.

The results looks very interesting, to the point where I am now considering adding TPI-like DEM filtering in my own code for much smoother contours that can significantly reduce the amount of contour redrawing that's needed.

The key is to be able to compare the raw and filtered contours on the same screen, I use a special "1m form line" object type for the raw contours so that the normal form line is still available when I decide that I need to add some of them, and then I can just hide those raw contours.

It would be cool to hear what is your impression of KP 1m contours with default smoothing, compared to yor approach.
https://twitter.com/RouteGadget/status/97858869235...
I tend to make basemap contours like that but with 0.625 or 1.25 interval.

I'll see if I can try that tomorrow.

As I said I really don't trust machine-generated smoothed contours, which is why I'll insist (at least for now) on having all the raw contours as well.

Last year I was in contact with an American orienteer who experimented with alternative DEM smoothing approaches, what he said worked best was a 2D filter which effectively meant that the degree of smoothing was large along the contour direction and much smaller along the fall line. The effect should be similar to the TPI blending but less likely to remove small reentrants (because the contour direction would vary).

Optimal contours should probably gravitate towards areas where the second derivative is high, i.e. bottom of hills and edges of hilltops.

I would not trust any automatic smoothing process for our gold mining terrain.

That's why you shouldn't trust it! Just use it to reduce the amount of manual work needed. Even in gold mining terrain (at least the maps I've seen via WorldOfO) has parts that hasn't been all dug up, so the contours look a lot more natural.

Terje

What I think I read in the FAQ at OCAD was that using the discount means you basically loose control over your current license. So, if you let the subscription laspe, you won't have any OCAD--other than a viewing version. Plus, I think I read that the 40% discount (if you have an OCAD 12 license) was just for a one-year subscription with the discount being 15% for a 3 year subscription.

You'd need to read it yourself to verify this but, given that, I will probably keep my OCAD 12 license and just get a new subscription if I decide to go that way. Than, if I don't keep the subscription current, I'll still have OCAD 12.

Within a month or so of purchasing OCAD 11, OCAD 12 was released and I could not justify shelling out another huge chunk of cash to upgrade. Now I regret not updating to 12 before this subscription nonsense was introduced.

I don't think I, or our small club, will ever update OCAD again - we cannot justify or afford that expense every year. Additionally, the concept of being held hostage to paying an annual fee to maintain our maps is abhorrent.

Not upgrading will cause problems down the road as contract mappers will eventually lose the ability to export to OCAD 11, but then we'll just specify that we want OCAD 11 files, or we'll move to different mapping tools.

I suspect that this thought process will repeat throughout the O world, to OCAD's detriment. As has already happened to other software systems that have moved to subscription models, it will spawn/encourage competitors that are sensible.

The best offer (by far!) for OCAD 2018 was the one where you bought OCAD 12 during the last few months, while 2018 was in beta.

OCAD have always allowed a free upgrade for new licenses bought during a beta period, now the offer was to get a free 2018 upgrade for three years, but then if you didn't renew after those 3 years, it would revert to a perpetual OCAD 12 license.

I.e. you would not lose control of your ocad 12 copy.

I suspect/hope that the same goes for anyone upgrading from 11 or 12 to 2018, i.e. the old copy keep on working.

Terje, as Carl pointed out above, if you upgrade from 11 or 12, then you will lose the ability to use your old copy of OCAD when your subscription period ends. From the OCAD FAQ section:

"What happens to my old (<= OCAD 12) OCAD version?

It depends. If you have purchased an update to the new version, then you lose your right to use the old version after the end of the subscription period (see terms of delivery). On the other hand, you get a discount when you buy an update version. If you buy a new version, you can continue to use your old version."

Another way to think of the offer to upgrade from 11 or 12 to get the discount is to view it as *selling* your (perpetual license to) old copy of OCAD for the discounted amount.

As an example, I currently use OCAD 11. If I were considering a 1 year subscription of OCAD 2018 to give it a try and see how I liked it, I could upgrade from 11 for about $115 USD or get it without upgrading for about $145 USD.

OCAD 11 may not have all the latest bells and whistles of OCAD 2018, but it's a hugely useful and powerful program as it is, and does everything I want and need today, There's little reason to think that will not be the case for years to come. To sell my right to continue to use this software for as long as I want for a pretty trivial 30 bucks strikes me as crazy.

Other people will bring other considerations to the table in considering the trade-offs and may well decide differently, which is fine. It is probably true that a good many users will prefer always having the latest upgrades automatically, as will be the case with the subscription model; at least that possibility would not surprise me. And so they can switch to 2018, taking the discount, and never look back. Nobody is holding a gun to their head.

Speaking more generally about the cost of the upgrade, OCAD does also address that in the FAQs, and the way I read it, they are saying that a 3 year subscription costs about the same as upgrading from older versions every three years would have been:

" The subscription option for “Single Users” is slightly more expensive for one year with much better update and support services, for 3 years the price is the same as the previous OCAD versions."

I can't say how the prices compare for users working in other currencies, but here in the US, working with USD, that would be my experience also--that a 3 year subscription would work out about the same as upgrading to the most current version every three years in terms of cost.

Stepping back and viewing this subscription model cost more generally, you can ask the question whether it's reasonably priced or a blatant rip-off. Thinking about about a mapper or a club working on just a single "typical" map project a year--which I will define as a map to be used for a ranking event, and about 10 km2 in size in average mapping difficulty terrain--how would the cost of OCAD compare to the overall project cost, whether measured by time used or cost? Well, I don't know what the latest figures clubs in the US are using for bringing in professional mappers, but a conservative ballpark figure for survey and drafting (before extraneous costs such as lodging, gas, etc.) is probably around $1000/km2. And such a project could easily take about 400-500 hours of effort. Put almost any amount of value you care to for the time, or take a rough estimate of project cost of $10K, think about how many hours you will be working with the software, and $150 for the year subscription is hardly worth thinking about--1.5% of project cost, and free for any other use for any other purpose the rest of the year.

For someone who doesn't need the latest and greatest and already has a working copy of OCAD 11 or 12, or for whom Open Mapper will do, then fine--use that. But that doesn't make OCAD's subscription cost an exorbitant expense.

@Swampfox: I agree completely with your conclusion, i.e. that OCAD (any version) is a tiny part of the total cost for any serious mapping project.

I.e. the pro bono base mapping work I did for WOC2019 in Østfold has been valued at NOK 500K (about $65K), since that was the best offer from a professional mapping firm.

I did know about the way you lose your current perpetual 11/12 if you upgrade, that's why I said the best deal they had was the v12 bought one or two months ago, during the 2018 beta, since that would in fact allow you to upgrade to 2018 for free and still revert to v12 after the end of the 3 year period.

I completely agree that was a really good deal. I wish I had known about it! (I must not have paid enough attention to OCAD news that gets published periodically.)

Fortunately I don't need any more than OCAD 9 for the mapping that I do. I can't map bush.

When I start 0CAD 11, it always mentions that there's a service update, but I've always been intent on getting something done, and have always skipped that. However, with the upcoming changes, I figured I ought to install the update while I have the chance. I started it up and this time clicked "Install Update", but that just took me to the 0CAD website. I went to Downloads-ServiceUpdate in the menu there, and it took me to a page that lists the updates for various versions, including 0CAD 11, but that says to contact 0CAD support. There's a link for 0CAD support, but I browsed around over there and the only thing I could find that looked relevant was a link that pointed me back to the Downloads page. Am I missing something? (Yes, I am missing something. Can somebody explain what I'm missing? Did I miss my opportunity? I did finally find an email address and sent them a message asking for the update package.)

@tRicky: The one feature I really miss when I downgrade a map to ocad9 is point object heights.

At least on 11 and later both line and point objects can have an associated height, I find that to be very nice when looking at dot knolls.

For paid mappers, Swampfox's calculations make sense. For a club, using free tools may be the right, easier answer in many cases. Let a club expert handle the basemap stuff, and let volunteer club mappers doing a local park download a free tool for their platform(s) whenever they like, no license tracking, no pressure to get done and release the license, plenty of freedom to just putter, even if the result is just improvement of their O skills, or a small map of a small park after some years. I set a 13km course in 0.7 sq km, and had someone wanting to continue the course at the event closure. Letting numerous members participate in mapping has a positive effect on skill progression and on the number and variety of (often small) orienteering maps, which keeps the sport fresher. The New England Orienteering Club had fifty maps, mostly small, some big, when I was last a member, and also, probably partly as a result, lots of members and lots of champions at all ages.

If I were mapping full time again, I'd have few qualms about paying hundreds a year for tools. If I were a club mapping director again, I'd lean toward free tools, and less stress, for most volunteers and projects.

Gian-Reto sent me the link for the service update, so I'm all set. (But I am wondering if OCAD is going to present an advertisement for the new subscription version every time I start the program.)

I believe that OCAD makes most of their revenue not from the Orienteering community but from Commercial (city) mappers.

I believe "Swampfox" is talking about the basic version of OCAD and not the Professional version which is much more expensive.

I myself find OCAD much better than OOM and KP. My biggest problem is field checking with my GPS unit. As I get more experience this should become second nature. OCAD 12 does not allow me to download/import areas, so I have to import them as curves and convert them to areas.

The biggest problem with older versions is that they don't handle georeferencing seamlessly. As we get more and more lidar of places we want to map, having solid, up-to-date tools is going to be very important to save time.

Higher resolution aerials are a thing now. The files are big. 32-bit software can't hack it without hassle---either cutting up the aerials into tiles or reducing resolution (for example creating 12-inch aerials from 6-inch aerials (that's 30cm from 15cm), which is throwing away nominally 75% of the information in that aerial (1 pixel in the place of 4). That's a real hassle for a mapper whether they're a pro trying to make money or an amateur with limited free time.

The biggest change I'm seeing in amateur mappers is the desire to use an iPad or Android tablet with or without a bluetooth GPS. Using OCAD in the field typically requires a pen laptop, such as an Panasonic Toughbook with Windows and a stylus.

Have you found tablet mapping to be effective yet? I've found it hard to use the drawing tools in the field, negatng the biggest potential benefit (of being able to render the map exactly while in the field in order to validate the rendering, spacing, depiction before moving on).

I've not had problems with OOM georeferencing, nor with it showing a detailed aerial photo (detailed enough that most bushes and rocks it shows seem to be too small to map). Has anyone else?

@dofishman: OCAD 11 and 12 can both import areas, at least when using DXF files as the source!

I worked quite a bit with them getting this to work reliably, both import and export, and in particular we found the proper way to specify islands.

I.e. any kind of area with holes in it is specified in the form of a surrounding closed polyline, optionally followed by one or more internal (also closed) boundary lines.

OOM made sure that they could handle the same DXF specification, so I have been importing the exact same Norwegian topo data to both OCAD and OOM. (I had to write a SOSI to DXF converter to handle the special format our mapping authority uses.)

@cedarcreek: All versions of mapping tools that should be considered for O maps have to hande georeferencing properly!

In fact, even when you take an OOM map and downgrade it to OCAD 8 so it can be moved back to OCAD, the georef info does stay there so you can switch the map back from local to georef coordinates and it still works.

I have bought a couple of WIndows Surface machines in order to try to use them for mapping, but I still tend to do most survey work with an A4 printout in the largest scale that will fit the area I'm looking at that day.

I draw in the stuff I see while hitting the split time button on my garmin watch. Immediately after I get back inside I download the track log via Quickroute and convert it to a GPX file, since this converts the split time positions into waypoints. I then import this file and use the track log (imported as a very thin red line + control rings for the waypoints) to verify the surveying.

Terje. Using OOM or OCAD for the surface?

@TIL: Either of them work! I.e. as I wrote I worked with OCAD to get areas with holes to import correctly, and OOM at the same time made sure that their code would do the same.

Last time I tried OOM on the surface the pen wasn't communicating with the software. That must have changed.
Update. They have. Path ahead now clear.

@Terje: While the georeferencing does stay there when you "Save as" OCAD 8, it's still not as easy to use as later versions. For example, if you want to load a lidar raster image or an aerial as a template, OCAD 8 makes it difficult to switch easily, while later versions accept a lot of templates ("background maps"), and let you quickly look through them.

OCAD 8 accepts only GeoTIFFs and doesn't accept world files, so JPGs and PNGs can only be F9'd.

OOM, in my experience, doesn't accept GeoTIFFs unless there is also a world file, but as long as there is a world file, it takes JPGs and PNGs.

OCAD9 doesn't like PNGs, I believe.

My standard work is to provide only GeoTIFFs with world files for OCAD 8 and 9 users.

GeoTIFFs are the lingua franca of geospatial software, so I find it convenient to just output everything as a GeoTIFF, even though the files are "BMP huge". In QGIS, I can output as a GeoTIFF with JPEG compression, but some cartography software doesn't accept that format.

I make basemaps much more than I map, and it's always a relief when know the mapper is using OCAD 11 or better (maybe OCAD 10, too?), because I know I don't have to take extra steps for backward compatibility. But the extra steps aren't all that much work, and it's not a terrible hassle, even for OCAD 8.

Ocad 8 has live GPS while 9 and 10 standard doesn't you, had to have pro. It may keep some using 8 if not yet upgraded all the way to 11.

OOM will shortly also be able to export to new OCAD formats. There is no GeoTiff support yet but it should be quite trivial to add. Creating a world file in QGIS is very straightforward though you just go to "Raster -> Projections -> Extract Projection..".

@mattp: That is _very_ good news, it will make it much more feasible to have a huge OCAD map database and export/import parts of it to allow new mappers to use OOM. :-)

Yes, as a GIS expert but OOM beginner I certainly see the benefits of using QGIS as a complimentary tool, I don’t use OCAD but know OOM pretty basically. Would be great to see it developing even more.

I was wondering when cloud-based 3D Lidar technology will be used in preparing orienteering maps, seems like an important development.

Free tools include OL Laser, KartaPullautin, OOMapper, LAS tools, and QGIS. To entice the mapper to make a purchase, OCAD must provide some additional capability that is not already available. So far, the only advantage I see is having everything in one tool.

LASTools aren't actually free. You can use them without paying, although there are limits imposed which cause me much grief. I've been meaning to suggest we approach Martin about a non-profit or other reduced rate.

gudeso, what do you mean by cloud based 3D lidar? I've certainly used free lidar data available on the web for an orienteering map, processing it by the free kartapullautin tool. I think that many are using free lidar, processed by various free tools. Do you mean something else?

What I want to see happen in the future is for orienteers to start using tools like QGIS more heavily rather than rolling their own lidar processing packages. Tools like QGIS are what professionals use to do this sort of analysis and it is fairly trivial to write extensions with nice interfaces to generate the same sorts of cliff maps, contours, smooth contours, vegetation and so on.

Then it is much easier for other people to build on the work and for people who are not on windows to use people's efforts as I have found using all base map generate programs on OS X problematic for one reason or another.

The learning curve with QGIS is much steeper than, for example, OL Laser. Once you are over the learning curve, fine. But most part time mappers can get where they want without going through that learning curve. You might just have to buy windows.

The last time I tried to use QGIS to process LiDAR data .las or .laz it always bombed. I did with difficulty get it to process a .adf LiDAR data file. From what I know QGIS requires LAStools to process LiDAR data and it will not handle a full tile like OCAD 12. In fact I could never get it to process even my split (small) LiDAR file even after following Martin's directions.

To me the learning curve is too steep! If you have a step by step set of instructions we can all follow that would be a great help to this community.

QGIS is more of a front-end for LASTools or Grass GIS. I'm pretty sure it doesn't natively process lidar.

The orienteering software is the first stuff on the block, so to speak, so in some ways, it's the only game in town. Orienteers are doing stuff on the fringes of lidar anyway, in my opinion. Our maps are more detailed than almost any others. Archaeologists seem to be doing very highly detailed lidar processing for very specific areas.

I do find QGIS is an essential part of merging, reprojecting (warping), and clipping aerials to make files that are of the exact area of interest, rather than using memory to hold the "available tiles" rather than just what you need for the map.

OL Laser was effectively useless for larger maps due to scalability problems, and even after Jerker introduced batching, those batch functions usually crashed far too often.

LASTools has a significant learning step, but the code has been very stable indeed, and it is purpose-made for large-scale parallel processing.

I love OL-Laser for quick raster outputs of large areas. My biggest complain is the lack of bounding box functions so tiles merge seamlessly. If that is fixed, I'd love the ability to output and merge the Grids (DEMs) from the batch process. I only have trouble with OL-Laser if my tiles are too big. I can't create 1m Grids if the laz files are bigger than 40-60 MB. I go through the cumbersome process to set up a batch even for one file. I'd love the ability to save the batch setup and just specify the input files and output directory details. I have almost zero trouble with OL-Laser.

I definitely need to learn more about chaining LASTools commands to create complex batch outputs.

@cedarcreek: Good to hear that you have better experience with OL-Laser!

I typically work with LiDAR data sets in the 300 MB to 3 GB region, so I have to tile them heavily. I always use solid overlap/boundary regions between those tiles so I don't have any issues at all with the final merging process.

In my current workflow I start with 256x256m tiles, with a 32m buffer, then for any tiles that are still too big I subdivide them into four 128x128 (+32) quadrants. I even do a final stage where those 128x128 blocks are split into 64x64 plots, but at that stage the buffer covers 3 times more area than the core, so the overhead becomes quite significant. I could probably reduce the buffer size for tiles with so dense point clouds, but that final split is rarely needed and the logic is much simpler with a constant buffer.

If someone wants some general cartography training with ArcGIS, I came across a link to this free seminar.
https://www.esri.com/training/catalog/596e584bb826...

A free license would be good with the free seminar. ;-)

I believe there is a nonprofit or personal home annual license for ArcGIS for about $100 per year. (I just use QGIS, which is free and open source.) I know a guy who pays for the license, but it's because of the extensive map files available with that license. He's a researcher for lost soldiers (MIA/KIA) for many different wars.