Subscribe to this thread
Home - General / All posts - Exporting to AI with Release 9
#28-Dec-18 01:17

After searching everywhere in the manual, online and in this forum I cannot see how to get a layered Adobe Illustrator file out of manifold R9 in the same way I have been doing in R8 for some time, ie. from Layout - Export, image, Ai files (*.ai). Does anyone know if this function still available in the new version?



#28-Dec-18 04:39

Further to above, it seems the export layout function shown below (from the R8 documentation) is no longer available in the R9 version?



8,743 post(s)
#28-Dec-18 04:47

As far as I know, there are no visible plans to support .ai as an export format from 9.

But I really don't know.

If you need or want it, you should definitely request it.

Bear in mind that Manifold 9 is not an iteration, but a new product.

For my part I will be happy enough with DXF as an interface to Illustrator (though that is from drawings, not layouts, and yes it is primitive--on the other hand almost unbreakably simple).

I had little success using .AI format from 8. It sounds if you have got it working better.

#28-Dec-18 05:36

Thanks for your reply TJHB.

I realise 9 is a whole new product but I guess I was hoping more for an improved, faster and more powerful version of R8. Or at least similar export options...

As a cartographer, virtually every project I have started in Manifold has ended in Illustrator and yes, I found the layout export/image/ai option both efficient and very useful for this. The thought of exporting one drawing layer at a time makes me shudder (as seems to be the only option for QGIS as well).

Nonetheless, I have bought 9 now so will continue with both versions, submit a request and see what develops in future. Not like I paid Arc prices I guess!



8,743 post(s)
#28-Dec-18 05:43

Don't despair. The two facts that Manifold does support DXF, and is very highly programmable, mean that interacting cleanly with Acrobat is fundamentally possible. (And in a way, easy.)

There is no need to export one drawing layer at a time. Whatever is most efficient, we can do.

Please keep in touch? Specific challenges would be very welcome. To repeat: we can do whatever is most efficient.

#28-Dec-18 06:00

OK will do.

I'll have to look at the print to PDF file via layout option too. Maybe the map layers will come through there into AI.

Thanks again.



1,918 post(s)
#28-Dec-18 10:01

In M8 I have always exported as pdf and loaded that into Illustrator for any further cartography.

Aussie Nature Shots

#29-Dec-18 03:37

Thanks ColinD. Not sure what kind of cartography you do to a flat PDF out of Manifold, but I would not consider doing anything even half serious without layers. Some of my maps end with 50 - 60 of them.

In M8 it's annoying enough that the layers are all within a clipping mask that needs to be released before editing and the irrelevant bounding boxes deleted, as well as converting the colour space from RGB as it seems Manifold does not support CMYK colours - the only way to print properly in wide format or offset. However, I have learnt to live happily with all that, knowing that I can at least get the layered AI file out, unlike in M9...



8,743 post(s)
#29-Dec-18 04:38

I reckon...

It would be very much worthwhile (yours and mine and...) if you could list every single step you apply with Manifold 8, to export a layout for use in Illustrator (with version).

I don't know if anyone has ever taken the trouble to do that before. (Yes that's odd.) Could you?

If we are lucky, this might help shape Manifold 9 design, substantially.

#31-Dec-18 04:13

Hi TJBH - it's not an overly complicated process, so I can summaries it easily with the addition of a few screenshots (wouldn't fit here so I have attached in a PDF for steps 1 - 3);

1. Import, sort, classify and manipulate etc all required data and create a map with layers. Note the simple, ugly RGB colour scheme and the even uglier label placement

2. Create a new map view at the required scale and projection. Create a layout from the map layers, set the page size and orientation (need to choose the Adobe PDF driver from the print option to access page sizes larger than A3), then set the layout properties as per the shot below - most important is the scope as 'view'.

3. Go to File, Export, Image and set the file type to 'AI Files (*.ai)'. Choose a location to save to and settings as per below;

4. In about 3 seconds your AI file is ready to open in Illustrator and begin the cartographic tidying up to turn it into a useful, legible and nice looking map.

5. If there are more than a couple of layers of labels, then I usually do separate AI files for each and then combine in Illustrator as Manifold has no way to set rules to manage collision avoidance of labels - it just leaves some off and this happens a lot when there are features close to each other, even on the same layer. For this reason I also set the point size as small as possible and make the map scale and page size as large as possible as this gives it more breathing space to hopefully label everything I need. Label size and style and map scale are all easily adjusted in AI.

Hope this is what you were looking for and perhaps useful?

AI export process_Manifold.pdf



5,436 post(s)
#31-Dec-18 05:45

Great thread! Some notes...

and begin the cartographic tidying up to turn it into a useful, legible and nice looking map.

That's the key... what do you do in AI that you cannot do in 8, but would like to do in 9? A good example...

and then combine in Illustrator as Manifold has no way to set rules to manage collision avoidance of labels [...]

Label size and style and map scale are all easily adjusted in AI.

... the example being, "I can do better labels in AI." That's good, but to turn that into a specific set of "this is what I want in 9" we need to drill into that with more detail.

First, Release 8 does have ways to set rules to manage collision avoidance of labels. See the options in the Labels Display Options topic in the Release 8 manual. Are those rules everything that everybody wants? No, of course not. But it is better to start with a consideration of the controls that already exist, in 8 or in other systems such as the Maplex engine in ESRI, as a foundation for what might be done better in 9.

Second, what is missing in the phrase "are all easily adjusted in AI" are the key additional words "if you want to spend time doing all that manually." Sure, AI lets you manually adjust label size and style, on a per label basis. So does 9.

That is where you get into some of the differences between a GIS, which has to answer for any changes made to the data, and AI, which doesn't give a hoot about the data, couldn't care less about keeping, say, city points georegistered for later use, and is happy to trash your data forever in service of getting a better graphics art effect for this moment.

I'm not minimizing in any way the value of beautiful and useful graphics artistry: I'm just saying that the AI way is very much "one way" in its total lack of awareness that data might be valuable in the future. However you want your GIS to be better at display, you don't want it to take the "let's make this simple by throwing away the valuable information content of our data" attitude that graphics editors have.

We also have a much more demanding task in GIS than AI sets itself. The AI way is to have many controls for graphics arts that are manually applied, for the most part, so a diligent artist can invest creative skill to produce a visually attractive display. In GIS, we don't want to use the manual word when there are a million objects in the display that need to have optimal display. For us as a community, the word we want to hear is the automatic word. :-)

One reason that is true is that GIS views change in a fraction of a second as you pan around, zoom and so on. It's not a static situation like a composition in AI, which still retains virtually throughout the original task of creating a printed illustration. GIS views assume that even if it is not a fully interactive display within the GIS console, the destination for the published view is the far richer and more demanding world of digital consumption: people may be viewing the thing on a smartphone, on a tablet, on a single monitor, on multiple monitors, through various web sites, within pop ups and so on. It's not a one view suits everyone situation.

Our task as a community is to express what we want in terms of flexibility and end goal for all of those situations, a classic, finely-artistic, static, printed presentation aimed at a fixed paper size and scale being one of them. There is no reason we cannot have it all, or at least in a form which automatically brings us very close to what we want in the various situations so that a limited amount of manual tinker time can get us exactly what we want in any given situation.

With AI, exactly those features which make it great at that one situation, a classic, static, printed presentation, make it lousy and inflexible in all the other situations. Given the chances that Internet is here to stay and that more sophisticated presentations designed for the web are likely to continue being popular (nobody is abandoning the web to go back exclusively to paper), we have a good opportunity now with 9 to craft something which is very useful both for the printed world of yesterday as well as for the digital world of now and the future.

For all the wonderfulness of AI, I don't see that same opportunity for AI. That's another strategic reason to put emphasis into effort into improving the already-rapidly-evolving feature set in 9, instead of into speaking Sumerian to AI. :-)


8,743 post(s)
#28-Dec-18 20:27

I'll have to look at the print to PDF file via layout option too. Maybe the map layers will come through there into AI.

No layers, since via a printer driver, by its nature flattened.

That's the only current vector to PDF. There might be a separate export to PDF, with layers, eventually. See here. I would hope that an AI export might surface at the same time.

#29-Dec-18 03:40

Right, thanks. I will put in a suggestion and (fingers crossed) hope so too! I'm no expert but I think maybe there's not a massive difference between PDF, PS and AI.



5,436 post(s)
#29-Dec-18 07:40

I'm no expert but I think maybe there's not a massive difference between PDF, PS and AI.

I see you haven't googled the question. :-) That's a bit like saying "I'm no expert, but maybe there's not a massive difference between a lump of coal and a diamond.... they're both carbon, right? " :-)

PS or PostScript is a programming language designed to produce printed pages on printers. When it is used to make stuff appear on a monitor, that's because the window is running an interpreter that pretends to be a printer.

EPS (encapsulated PostScript) is a limited form of PS.

AI format is a brutally simplified subset of EPS, with the addition of many differences.

PDF is yet a different display programming language that, unlike PS, does not contain program flow commands, but which is unlike AI in that it is closer to PS as a language than what is AI. It's as if the programmer who created EPS got drunk and accidentally deleted all his sources, and then tried to re-create what he or she did, with some memory of how he approached the task but then writing it all from the ground up again. So, sure, lots of resemblance but not the same thing.

The problem with all of these approaches is that they were created to get around hardware limitations that were a big factor 30 years ago but which now are long forgotten. But still, they persist because of the large installed base of Adobestuff. Even Adobe itself has trouble sorting all this out.

As display-generating languages, they have some advantages. But as means to generate reproducible vector data they are absolutely, horrifically terrible compared to more sensible means.

By the way, for those who are puzzled by what I mean by a "programming language for display," consider two different ways of storing data about the exact size, shape, and location of a real estate parcel:

Store the data: Start a file by announcing the coordinate system you are using, using an EPSG code. Now, give a list of each coordinate that defines the boundary of the parcel in order, using the usual GIS conventions for islands and inner holes. Done.

Store a description: This is the old "metes and bounds" approach going back to Roman times. Instead of storing exact coordinates, you store a verbal description, such as... "Start at the rock where the corner of Farmer John's property touches the corner of Farmer Fred's property. Proceed south by southwest until you see the big oak tree exactly to your left. Put a rock there to mark the spot and then turn south and proceed 20 paces. Put a rock there and now turn due east..."

That's a programming language description using commands like stepping 20 paces and using program flow control logic like "until you see the big oak tree..." To find out what the parcel is you have to "execute" the "program" by following the instructions in the description.

The coordinate data is more precise than the descriptive way, because maybe not everybody following the programming steps has the same idea as to which of many oak trees is the "big" oak tree, and maybe not everybody agrees on the exact distance that a "pace" should be. That's why different programs which execute any of these ancient programming languages so often come up with different results.

AI is really guaranteed to work only within Illustrator, and at that only within those Illustrator versions that match the AI version, because it depends upon the peculiar internal assumptions made by AI. It's not a good interchange format and Adobe has never pretended that was the mission for AI. Adobe intended AI only to solve a specific set of issues 30 years ago, in a setting where they had to get Illustrator going to provide vector capability that was totally lacking in Photoshop.

It's a terrible hack that persists in the modern world, like some ancient snippet of virus DNA that's become embedded in the human genome, ready to wreak havoc should cancer or some other mishap throw the "on" switch for that sequence.

As for supporting AI, it would be far more effective in terms of providing ease, a great life, and wonderful capabilities to simply provide those vector editing capabilities in Manifold that Illustrator has, which you need to use. Illustrator is not so good at handling large data, so it's not like you will not have to find an alternative sooner or later.

It's like you have a crew of guys who are building houses for you and one of them, an older guy, is a first rate electrician. He does all the wiring, but he only speaks Sumerian. You'd like to keep using him, because he really is a great electrician, but it's a big hassle teaching all the rest of your crew and your architects to speak Sumerian so they can talk to the guy. No matter how you try, it turns out Sumerian isn't a particularly good language to discuss complex wiring for modern housing projects (no words in Sumerian for "electricity" or "WiFi") so there are lots of errors in the work.

All the rest of your crew are young, very smart workers who are super at using smart phones and taking instruction from the automated processes your business uses. But they don't know how to do wiring, at least, not yet. In the choice between teaching the entire crew how to speak Sumerian or picking out a few of them and teaching them what they need to know about doing basic electrician tasks, it's smarter and more valuable to teach them the basics of being an electrician.

hphillips16 post(s)
#29-Dec-18 16:32

Dmitri wrote "As for supporting AI, it would be far more effective in terms of providing ease, a great life, and wonderful capabilities to simply provide those vector editing capabilities in Manifold that Illustrator has, which you need to use. Illustrator is not so good at handling large data..." I could not agree more!

Export to Adobe Illustrator (AI) is inefficient and I agree that AI bogs down with large data like contours - plus there are expense, alignment and redundancy considerations. If on subscription the AI software expense is $400 annually. And there is the problem in AI to keep imported data aligned and scaled because AI doesn't work with GIS coordinates. Many cartographers alleviate the cartographic/GIS limitations of AI with the addition of the Avenza Map Publisher plugin. Over the years Map Publisher has gotten increasingly complicated with more and more GIS capabilities. That was accompanied with increasing cost - a Map Publisher non-expiring license is currently $1400 with $350 yearly maintenance. Avenza has done a good job to add cartographic and GIS capabilities to AI; unfortunately small distribution and continued development costs to keep up with new versions of AI probably justifies their software cost to them. It's crazy though - building all that GIS capability into a plugin that only works inside AI and is redundant to a more capable external GIS. The AI / Map Publisher ride was too expensive for me and I got off at AI6.

Going outside Manifold for map finishing makes the process complicated and expensive. Manifold has indicated plans to improve the cartographic capabilities of its GIS. When AI / Map Publisher can be bypassed using Manifold alone it will be a welcome day.

So returning to the thread subject by JohnF (nice work at your website BTW)... that you as a cartographer indicate you usually finish your projects in Adobe Illustrator indicates some missing desirable capabilities in Manifold. What cartographic map production operations do you do in AI that you cannot do in Manifold? What cartographic capabilities does AI have Manifold does not? You mentioned RGB/CMYK color space. Better type control - tracking/spacing leading? It would be great to see those operations/capabilities listed here and submitted as suggestions to Manifold development.

#31-Dec-18 01:21

G'day H.

Thanks for your comments (and praise) above.

What cartographic map production operations do you do in AI that you cannot do in Manifold? What cartographic capabilities does AI have Manifold does not? You mentioned RGB/CMYK color space. Better type control - tracking/spacing leading?

Yes to those things and a heap more, probably too numerous to list here (and as mentioned above I don't want to bother the Manifold team with all of them anyway).

But the single biggest one would be type (label) placement. ESRI has been working on refining cartographic output with respect to auto label placement for over a decade with the inclusion of Maplex in Arc Map; and Map Publisher about as long with Label Pro for Illustrator, and I'd suggest both of them are still lacking in finesse compared to any manually finished map out of AI. That's just one of the major reasons I still go there.

I don't know about M9, but the label placement in M8 is pretty ordinary, and that's just for point features. For linear objects (even straight roads) it's truly horrible.

I'd also point out that I distinguish GIS 'map makers' quite distinctly from cartographic map makers, such as myself. Much of the output from GIS operators is probably quite adequate for the purpose required - it just doesn't stack up for most of the purposes I create maps for.



5,436 post(s)
#31-Dec-18 06:33

probably too numerous to list here (and as mentioned above I don't want to bother the Manifold team with all of them anyway).

A big mistake on two fronts. First, it seems the features are mostly manual editing ones and there aren't too many of them, probably no more than a few dozen, tops. Certainly less than two or three hundred. Yet, if you look at any of the recent announcements in builds, a series of builds in a month typically have a few dozen top level features with each resolving to many sub-features, often totaling in the hundreds (the sub-features normally not being listed separately, for example, things like clipping Left and Right effects automatically by terminal graphics in lines, the very many sub features involved in positioning terminal cap ends in various situations, etc.. The scale of the list is not very big at all. It's much smaller than the rather long list of requirements database people often send in.

Second, it's not a bother for either the forum or for the Suggestions process for you to say what you want. That's what the Cutting Edge process is for. Refine your thinking here and then send in a Suggestion. Can't get what you want if you don't say what you want, right? :-)

I'd suggest both of them are still lacking in finesse compared to any manually finished map out of AI. That's just one of the major reasons I still go there.

Ah, well, then if that's the case then the situation is very easy to fix. :-) Few things are simpler and easier to do than manual features. If people are willing to do the work manually, no problem.

The reason GIS doesn't do that, as a central approach like AI, is because GIS do not want to mess with labels and features one at a time. AI has no choice in the matter since it is pretty much not automatic-anything. GIS therefore has much bigger and more demanding expectations than AI: ESRI put all that effort into Maplex not because the ESRI people don't know how to cobble controls that force people to manually adjust millions of labels one at a time, but because GIS people demand something better. GIS people want the GIS to do all that for them.

Consider what life would be like in AI without any GIS: draw that sample map of the region around Pretoria, drawing each line manually, and then label each and every item from scratch, hand positioning and hand drawing every danged label. One. At. A. Time. ... ouch! :-)

There's also the consideration that GIS seeks to bring greater capabilities to a broader range of users, including those who have totally average artistic capabilities. One of the nice things about modern GIS is how it enables even those with little artistic talent to make acceptable displays.

It's true that people with poor taste can use automated features to create spectacularly ugly displays. I personally have thoughtfully contributed many examples of poor taste from my own work to illustrate the point. :-) But having automated style capabilities with some reasonable, pre-built options (like the Color Brewer palettes) can help guide people into better displays.

Last, but not least, what is striking about AI and Photoshop and similar is how ancient they are. Photoshop today is amazingly similar to Photoshop of 20 years ago. It's basically the same thing, and if you look at how experts use the thing they continue to use mostly the same subset of commands today that they have been using for the last 20+ years. That subset of commands is not very big, a few dozen at most. It's not hundreds.

So, if manual editing is the game, that is perfectly within reach. Name the top 20 commands and capabilities you use in AI and it is quite likely (no guarantees, but quite likely...) direct analogs could be provided in 9 as well.

#31-Dec-18 01:02

That's a bit like saying "I'm no expert, but maybe there's not a massive difference between a lump of coal and a diamond.... they're both carbon, right? " :-)

Haha, yes great analogy. Glad I added that qualifier at the start!

I really only mentioned that because another app I use called CadPublisher seems to create all three of those outputs effortlessly from layered DXFs out of CAD such as MicroStation and AutoCad. It's been around since the 90s and I still use this method for all of one client's work today.

The layered AI file that comes out of M8 is a legacy version, but it still works fine with all versions of Illustrator. I don't believe it's possible to create one program that will do everything that everyone wants in the fields of GIS, mapping (cartography), design and graphics, so I doubt there's much point in listing a whole bunch of features that 'could' be added to Manifold to achieve this. Just like I wouldn't suggest AI with Map Publisher (I run this also) is everyone's idea of the perfect GIS because I don't think that was ever it's aim.

It is meant to provide a way of getting georeferenced spatial data into the industry standard, vector graphics design app while providing cartographic tools to map makers, to make maps. Not to store, query, manipulate and exchange massive amounts of geodata, at least I'd never suggest using it for that, because yes it is slow for large datasets - but so is M8 compared to Global Mapper. I often do a bit of viewing and clipping of large datasets in GM before going to Manifold because it's just way faster.

Anyway, back to my original issue. As suggested by TJHB above, I'll send in a simple request to the sales team for the addition of an AI export to M9 and see what happens, but I don't want to bother them with much more than that because perhaps I am like your Sumerian electrician!



5,436 post(s)
#31-Dec-18 07:04

CadPublisher seems to create all three of those outputs effortlessly from layered DXFs out of CAD such as MicroStation and AutoCad.

Well, if your standards are no higher than what is found in a DXF then that's very easy to do. Dumb down the content of what is in Release 8 to the level of a tiny DXF and it's easy.

Doubt my word? Try it and see: how well does CadPublisher do at creating identical programming source code (what PS, PDF, EPS, etc., are...) in the various Adobe display programming languages when the content is the profoundly more sophisticated world of GIS data, in particular a mix of limitlessly, dynamically rescalable data from web servers, many-channel raster imagery and complex vector data?

How well does CadPublisher do at correctly rendering a PDF that incorporates a big background frame from a Google street map web server, correctly scaling the contents from 72 DPI (screen) to 4000 DPI (fine printer) and ensuring that the tiles pulled from Earth have the right scaling to show their labels in other than microscopic print? Can it correctly draw a 50 GB GIS-generated vector layer that uses symbology of relative effects such as drop shadows, border widths and so on to ensure that the symbols really do appear at the right point size whether the display device is 72 DPI, or 300 DPI, or 1500 DPI or 4000 DPI?

When the CadPublisher output is consumed by another GIS or graphics design package, how well is the original data preserved? Has it been rasterized into mush or are the original rasters and vectors preserved? Just guessing, but I bet that the CadPublisher output only seems identical in those packages that don't take as close a look at the details (often the case in graphics arts editors) as does a GIS.

I ask those questions to apply the Socratic method: by considering the answers to those questions it is easy to see that the CadPublisher solution to a difficult task is to simply not do the task; instead, it limits itself to a simpler task: take the very simple situation of a DXF and convert that. Ignore the hard parts of what people routinely do in modern GIS. :-)

I don't believe it's possible to create one program that will do everything that everyone wants in the fields of GIS, mapping (cartography), design and graphics,

Sure, I agree, but that's not the task. Don't let the theoretically perfect be the enemy of the practical good. The task is to create facilities that let most people do what they need to do.

It's OK to create fast and convenient means that do what people end up doing 95% of the time, and which provide less convenient but still workable means to do the remaining 5%. That won't be absolutely everything imaginable, everything that somebody, somewhere, might want to do, but it would be a heck of a step forward. Most cartographic beauty in daily printed map production is the result of a fairly small set of manual commands/editing capabilities being applied with good taste and artistic skill. It's not hundreds of tools with thousands of features. It's a very small set.

Keep in mind that AI is far from perfect, even within the subset of the task that it has claimed as its own. You're used to it so it's natural to accept the deficiencies in AI, such as wretched slowness with bigger data and a total fail in the "automatic" end of work for things like creating beautiful web sites that are dynamically reactive to users.

Oddly enough, you can see that effect in action when making a pairwise comparison between Release 8 and Photoshop. 8 has surprisingly many graphics arts features in it, which, even more surprisingly, in some cases are much better at doing certain tasks than using Photoshop. People who have applied years of learning to become intuitive experts at Photoshop won't know that, because they have not expended similar effort on learning 8, but nonethless there are cases where 8 is better than Photoshop at certain graphics arts tasks. (The mix of vector understanding and raster within 8 makes that possible.)

In the Adobe world there is a similar artificial divide between AI and Photoshop. Adobe's products have that vector/raster divide which makes certain things more difficult that could be easier if the product were both vector and raster and had similar richness in each. A GIS framework that does both is a natural place to grow more convenient capabilities that more Balkanized products do not provide.

Just saying, don't be satisfied with old technology, not in GIS and not in graphics arts, and don't limit suggestions to a request to speak Sumerian with the focus being on a continuation of frozen-in-amber limitations. Think big and go for what needs to be done simply and automatically, whether or not AI or 8 or anything else from the past can do it.


8,743 post(s)
#31-Dec-18 08:13

Despite having said that I'm happy enough with the ability to take data to Illustrator via DXF, I do feel quite strongly about this. I very much agree with John.

So, for a few days I've been drafting and ranking a list of critical, essential features that I would need Manifold 9 to support if I were to use it for cartography (true cartography, not GIS mapping) without recourse to Illustrator.

Well, what an interesting exercise! The list of essential features keeps getting shorter (and the "nice to have" list keeps getting longer).

I'm pleased I didn't post the list straight away, because of how much it has changed.

I will post tomorrow.

Can I ask John, and others who do cartography with Illy, to please do something similar?

If Manifold is really going to pull the plug on Illustrator support (say, without alternative export to layered PDF/x) then this becomes important. Though yes, we can get around it with scripted export to DXF. Or renew our expensive maintenance contracts to Avenza MAPublisher. Yikes--but good for them.


5,436 post(s)
#31-Dec-18 13:35

If Manifold is really going to pull the plug on Illustrator support (say, without alternative export to layered PDF/x) then this becomes important.

Well, there's no support for Illustrator per se, there's just support for PDFs and nobody is talking about dropping that. We've gone a very long time without anybody mentioning .ai format, but hey, if that's becoming a hot button, where is it in the Suggestions page written that Manifold doesn't listen to suggestions? :-)

As far as PDFs are concerned, nobody, to the best of my knowledge, has indicated anything but continuing support for PDFs. Discussing the ins and outs of why it is such a bizarre format does not in any way indicate lack of continuing support for PDF. There's a huge commitment to making PDFs perfect in Manifold as a way of printing, and it's hard to do that without as a side effect improving PDFs in Manifold for those who wish to use them in interchange.

Regarding .ai format, I can't recall any occasion in the past where Manifold has said anything other than what the majority of users prioritize and ask for will get done ahead of those things which people feel is a lower priority. That's a good approach I think should be continued.

Part of getting the most good for the most people is being aware that some things take disproportionately greater time to accomplish. It's the old deal of some marketing guy coming in with a list of ten items where the tenth one takes 50 times as much effort as the other 9, and may not be a particularly smart thing to do anyway. But he insists all ten must be done even though the user community, if informed of the tradeoffs, would far prefer that the same amount of time be spent on implementing the "reasonable 9" plus the next 40 items on the wish list instead of the unreasonable tenth item.

The nature of PDF/PS/EPS/AI is that such programming language descriptions are absolutely horrific and unreliable at interchange of significant GIS data. They are a black hole for resources, and no matter how much resource you invest into them there will be unreliable results because whatever is written into the file will be subject to interpretation by some execution engine within the receiving application that is not under Manifold's control.

Consider the extreme variability in what the display results are when the very same PDF is interpreted by various PDF display engines from Microsoft, Adobe and other vendors, even when it is an Adobe product that writes the PDF. Do you really think Adobe has not invested zillions of dollars into being able to write PDF in a "standard" way? Is it possible that Microsoft has failed to invest into reading PDF correctly? Both have invested big money into this, but the results continue to be variable because that is the nature of PDF.

I therefore think it is a fair question for users how they would like to see all this approached in terms of priorities, and to raise the possibility that maybe it would be smarter first to invest the same resources into implementing missing pieces for vector graphics arts editing within Manifold than it is to invest into chasing perfection within ai export to Illustrator. It's easy, of course, to do a simple hack like what 8 does. But beyond that, maybe it is wiser to spend the effort into better native vector capabilities. After all, those serve everyone, and not just Illustrator users.

By the way, regardless of whether you think the balance is one way or the other, I still say that if anyone is interested int this question the first responsibility of any user commenting on this, in total "community driven" spirit, is to provide a list of the top commands you find yourself having to do in Illustrator because they aren't in Manifold. If you can't list the top ten or 20, at least list the top three or five.

Happy New Year! Time to head off to the party. :-)


6,316 post(s)
#31-Dec-18 14:55

For me good cartographic facilities of Mfd9 combined with a good PDF export is essential. I have never used ai export.

But I made use of layered PDFs in talks activating additional layers while explaining. Doing this with the delivered final cartographic appealing product was convincing and less work, than to produce an animated presentation with extra affords in Powerpoint.

So for me PDF is by far the most important export format and end product besides screenshots for text figures and PPTs.

Mike Pelletier

1,599 post(s)
#31-Dec-18 18:44

Like Klaus, I suggest focusing efforts on internal cartography and good PDF export. Layered PDFs are very handy as Klaus pointed out.

For fancier printed maps, I'll occasionally used Canvas from For a long time they have a product that attempts to marry CAD, graphics, and presentation. A cheaper alternative to the Adobe line. Less capable but easier I think to use.

Buried somewhere in the Mfd 8 help document is a statement that Mfd wants to marry GIS, CAD, and graphics. That has always been my hope for Mfd and it seems Mfd 9 is on its way. Certainly the styling of vectors is now fantastic!

Graphics programs are all about allowing creativity to flow. That requires fast application of tools and generous undo capabilities. Currently Mfd 9 with it's database focus requires a save after editing a vector with no undos. It's fine for now but I hope that can improve someday.

I agree that we need just a small subset of the fancy tools from graphics programs. Things like applying transparency gradients across vectors is nice but not important at this point.

Heading forward I think labeling is the biggest issue to tackle. My hope is that it will allow setting up parameters for each label to be layered including how to handle conflicts, run the automatic version, and then have it seamlessly allow manual adjustments that persist (or not by choice) with other edits. I suspect this will be difficult and time consuming. Hopefully though the groundwork can be laid to allow this to steadily improve over time to be near the top in the business and low hanging fruit can be added now.

Adding text kerning is important to bump up the graphics capability.

Vector editing needs to catch up with clipping/combining tools. From what has been said, I suspect much of that is in the works. Also we need a COGO tool added that allows applying rotations and going back to modify segments along the traverse. This was never added to Mfd 8 and was certainly a big hole for a full featured GIS program.

Sounds like grouping objects, alignment tools, scale bars, and north arrows will be included with legends that is to be released in a build soon.

Once this is all added, Mfd 9 can replace Mfd 8 for general map making. Oh happy day! Then back to raster stuff and the awesome speed of Mfd 9.

Happy New Year all!

140 post(s)
#09-Jan-19 17:00

Hi Mike,

Don't forget about the Canvas GIS version which directly reads a multitude of GIS formats directly. The advantanges of that are that you can, for example, have a map in your poster and zoom into it or out of it, inside your fixed frame poster, you can add GPS data to your Canvas file, thematically map GIS data layers, add JPEG 2000 imagers, or ECW images. It's 64-bit, some level of multi-core aware, and a full raster vector illustation program with massive level of precision and coordinate space. Even supports Seg-Y data and DICOM. Perhaps Manifold users could use Canvas for their cartographic functionality? Canvas 2018 GIS version is $799 USD, and they do also have a subscription option as well. I myself use Corel Graphics Suite 2018, I've been using CorelDRAW since 1997 (version 4). I have always been able to have far more objects, larger paper space and more complex things inside CorelDRAW than Illustrator. Yes Corel still exists... all the Adobe fanboys/girls always say that, "Corel still exists?".

Would there be any efficacy in trying to use the SVG export for the vector data and bring that into Illustrator, or is there an even more limited functionality than AI? I'm assuming SVG isn't going to support layering.

I use TerraGo Tech's Map Publisher ArcGIS Plugin and it works great to preserver layers, even transparent layers without flattening to a single raster imager which GeoPDF export does from ArcGIS Desktop (32-bit). However, TerraGOs Map Publisher can't handle more than two data frames in the layout.


8,743 post(s)
#01-Jan-19 00:32

Here's my list of essential items which would enable using Manifold for all cartography, instead of Adobe Illustrator. They are in rough order of priority for me, except that the last item should arguably come first.

Over the last few days I have changed my mind about what should be on the list and in what order. Others would write a different list, possibly longer.

  • controlled Z order per object (within layers)
  • blending modes (between layers), including at least:
    • lighten
    • darken
    • overlay
    • multiply
    • luminosity
  • bezier curves (with edit handles)
  • optional bleed for layouts
  • standard RGB ICC colour spaces controlling images and styles, including at least
    • sRGB
    • Adobe RGB
    • Gray Gamma 2.2
  • gradient fill for areas (linear and radial)
  • area styles for text (possibly by conversion of text outlines to areas)
  • colour sample readout, showing
    • (a) active layer colour under the cursor
    • (b) blended map colour under the cursor (perhaps if Shift or Ctrl is down)

      without interpolation, but with an option to average over a small radius

      (one number per channel, including alpha and Z if any)

  • PDF/X import and export, with layers, embedded fonts and colour profiles, for standards-based printing and data exchange

Here is an extra list of items that seem to me not essential, but nice to have, again in rough priority order.

  • CMYK colour model (besides RGB), with at least two standard CMYK ICC colour spaces
  • import and export of Illustrator and Photoshop colour swatches
  • import and export of Illustrator symbols
  • multiple styles per layer/per object, with rendering order, transparency and blending modes
  • transparency between objects (in Z order) as well as per layer
  • lockable layers
  • clipping masks (shallow and deep)
  • opacity masks (shallow and deep)
  • variable width lines and area borders
  • gradients along lines and area borders
  • ICC colour space conversion

I think most cartographers (and all designers?) would move the CMYK colour model to the essential list, but in theory this can be left to conversion after export. On the other hand, it seems almost essential that on import from PDF/X, Manifold should be able to convert CMYK to RGB (respecting the source colour space). Instead it could, at a pinch, reject any PDF containing CMYK colours on import, but this would not enhance professional relationships.


5,436 post(s)
#01-Jan-19 05:53

Thanks, Tim! This is perfect, one of the best lists I've seen. Count me in. I will advocate based on this for a running start for the new year.

125 post(s)
#01-Jan-19 07:36

I use Corel Draw for cartography and it is not inferior that AI I think, or at least not so much inferior to justify the difference in costs. AI costs 30$/month and Corel Draw costs 700$ without need for a subscription. Corel is as mature as AI is and there is no reason to constantly upgrade. You buy it once and you are done. So export from M9 to CorelDraw file would be nice for me, although I don't expect it to happen, and it also not so essential. I actually don't mind much exporting to dxf and reformatting everything, since I don't think there could be a way to export a Manifold layout and get a almost ready Corel (or AI) map. You would always have to do much reformatting. Anyway, doing the cartographic stuff in Manifold is what I would prefer, so here are some ideas.

First of all from tjhb's list I find top priority:

  • controlled Z order per object (within layers)

  • bezier curves

  • conversion of text outlines to areas (not just to fill them with some fancier fill, but per se also, would solve some problems with printing companies)

  • layered PDF import and export

To add to the list, the most important would be:

  • Locking styles rendering (patterns/label sizes, dash patterns, linear widths etc) for specific scale for drawings and labels (must have to WUSIWUG)

  • Option to manually move/rotate labels at specific positions and lock them there regardless of pans and zooms.

  • Align and distribute in layouts

Nice to have:

  • Export layouts to .svg files. Although a limited format, it could work for simple maps if you want to edit in AI/CorelDraw

  • Hide layers parent vector behind label. Here is why: A common cartographic procedure is labeling of contour lines. We usually want contour height labels at some intervals (and not in every single line), we want the texts to be aligned to the lines, and we want the contour line behind the label to be “trimmed” a bit, for better legibility. So it would be nice to have to option to automatically hide the contour line behind the text while being able to set hiding padding. You can set a box for label but it hides everything under it and not just the contour line This is a very specific task, but very classic and usual. It has been has been discussed years ago in this post

I will add to the list if I think anything else (pretty sure there are things I forgot)

125 post(s)
#02-Jan-19 07:17

Some more "nice to have" ideas:

  • Separate rendering of items subcomponents. In street maps, when drawing roads of the same type, supposing you have a special layer with roads only of a specific type, when the roads cross you get this result of picture crossroad1.jpg. This image is created using a style with a halo. This symbolization in one case gives the false impression that one road has a bridge over the other and in the other case that one road “enters” to the other. The right way to show road would be the one in crossroad2.jpg. In order to get this result one creates 2 drawings, one with the black lines and one with the yellow and stack one over the other in a map. Manifold could give the option to render firstly all halos of the lines in drawing and next all the standard lines. This way you could get the right result without the need for a second drawing. Since there is a workaround without the need for the special option I am talking about I am ok with that. But, when designing a legend for that combined symbolization, then there should be the option to...
  • Overlap various drawings' symbolizations in the legends, one over the other. In the previous case you should have the option to overlap the symbolization of the drawing with the black line and the symbolization of drawing with the yellow line. This option maybe is more important than just “nice to have”, since it would be useful in many legend creation cases. The “overlapping” doesn't necessary have to be “smart”, since just the option to (accurately) place legend items one over the other would do fine.
  • Labels styles with management of spacing between letters.
  • Labels styles with options for distance over or bellow lines. See picture textplacement1.jpg.


140 post(s)
#09-Jan-19 17:02

Nice to see a fellow CorelDRAW user here!


562 post(s)
#09-Jan-19 02:35

Blending mode is really difficult to understand . i study during many days of testing under photoshop.

In which Gis user case it could be usefull ?. I. Dont use a lot ai ( i have last cs6 before adobe goes to clound marketing to justify thé new subcription model : pay Montly for many=all Adobe tools you LL never use ./ master, drop fireworks Macromedia tools .. Thank sérif affinity) . So need to understand more when blending mode (s tyle) need to be use !!

Corel can do things better Than AI but the number of plug in under AI is outstanding i write plug in but Also by number of driver (cut and embroidery).adobe support automation of gui like excel script do !

join image

"Because my dad promised me" ( interstellar ) but blackhole don't exist

best hardware with no ads focus on quality features price like manifold see xiaomi


562 post(s)
#09-Jan-19 03:06

I think people should think sérif vector tools rather than illustrator when compare price and functionnalities really needed . ( not test by myself but to test buy your own to sée if it fullfill ....) .What i am writing doesn t make sense since thé use of AI is mandatory because we want/need to use AI Gis plugin. The first comment that come to my mind after first but manifold ( 5 in think) was woaho labels placement is "rudimentaire" basic not efficient... And things dont move a lot from this time !!! Except édit label to go to a new line and unicode support !

When read Label suggestion . This make me think of typography tools!!

join image

"Because my dad promised me" ( interstellar ) but blackhole don't exist

best hardware with no ads focus on quality features price like manifold see xiaomi

Mike Pelletier

1,599 post(s)
#09-Jan-19 16:34

I've found blending tools useful for mixing aerials, vector land use classification, and hillshading. The land use classification can be converted to a raster before blending, so I'm not sure blending tools are need for vectors if that is another big effort to implement. I've never tried to really understand blending (assume it gets super complicated with various colors) but rather just try out the various options and see what looks best.

Really like the suggestions in this thread!

Manifold User Community Use Agreement Copyright (C) 2007-2017 Manifold Software Limited. All rights reserved.