The Omni Group
These forums are now read-only. Please visit our new forums to participate in discussion. A new account will be required to post in the new forums. For more info on the switch, see this post. Thank you!

Go Back   The Omni Group Forums > OmniGraffle > OmniGraffle General
FAQ Members List Calendar Search Today's Posts Mark Forums Read

Disappearing lines Thread Tools Search this Thread Display Modes
Originally Posted by DerekAsirvadem View Post
My experience with Omni is, they fix bugs, period. Invisible lines such as this, are definitely a bug. In my experience (seven years, many versions) with Omni, bug fixes are not subject to votes. There is no bug that I have encountered, that is still outstanding; they are usually fixed in the next minor release.
Well, as someone who has filed many hundreds of bug reports with Omni, I can tell you that there are long-standing bugs which they choose not to fix even though fixing them should be straightforward, and when pressed, the word from Omni is as Rob reports: "not enough customers are complaining about that for us to take time away from things that get more requests." I'm not talking about feature requests, I'm talking about bugs. I doubt whether you would ship a product that had an obvious, easy-to-demonstrate flaw such as OmniFocus for Mac not retrying the sync that fails when the laptop wakes up from sleep and schedules OmniFocus before the network stack is fully up, because you seem to take intense pride in your work. Does the official Omni response here strike you as reinforcement of your position that bug fixes are not subject to votes?

The auto-layout feature in OG Pro 5 has plenty of bugs, as Rob has amply demonstrated. Trying to work around them often exposes bugs in the Applescript support. See threads such as,, and come back and tell us you can't see what OmniGraffle is doing wrong!
Originally Posted by whpalmer4 View Post
Well, as someone who has filed many hundreds of bug reports with Omni ...
I have not reported 100's of bugs, probably 25. And I have experienced only one crash in seven years. And I re-iterate that I am a heavy OG user (light OO user).

... you seem to take intense pride in your work.
Yes. And thank you for noticing.

Does the official Omni response here strike you as reinforcement of your position that bug fixes are not subject to votes? [...] See threads such as [...] and come back and tell us you can't see what OmniGraffle is doing wrong!
I stand corrected. Omni is a much smaller company than I realised. Evidently (from the Omni responses on posts that you referred) they lump:
  • straight bugs (that should be fixed without discussion)
  • rare bugs (due to the feature in which the bug is encountered being rarely used, and therefore possibly worthy of discussion)
  • new features
  • enhancements & changes to existing features
together, and they prioritise the lump on the basis of votes. Disgraceful for a cutting edge s/w company.

And I posted a response to Brian directly in the "sync after sleep" thread, expressing my upset, and the conflict in expectations, and the consequences, so I won't repeat that here.

We now know what we can expect in terms of changes to the product, from a company that has a short-term, non-QA approach to their business.

But leaving it at that would be incomplete for me ...

How Important Is It ?

The auto-layout feature in OG Pro 5 has plenty of bugs, as Rob has amply demonstrated [...].
Ok. I read the lot (but I can't be asked to open the zip files, etc, and very few others have). For the life of me, I cannot understand what Rob is going on (and on and on) about. Yes, there is a problem with auto-layout when orthogonal lines are used; yes, it clearly works well with, and is obviously skewed for, straight lines; to go on about the resulting layout being "incorrect" and expecting (demanding ?) symmetry for orthogonal lines is way over the top. My considered opinion is that auto-layout with straight lines is fine; auto-layout with orthogonal lines while not being "incorrect", could certainly use some enhancement; evidently there is not enough votes, so forget about it.
  1. I am all for the symmetry; clear hierarchy; clear parent-child group separations; and the reasonable amount of intelligence required to lay out a diagram with orthogonal lines different to the layout using straight lines, but clearly the product does not have that, you are getting object placement carefully designed for straight lines, there is no different placement for orthogonal lines. Deal with it. In fact I invariably use orthogonal.
  2. In the time that Rob spent posting those several threads, pages, and accompanying diagrams, I could have moved about 1,000 objects in 100 diagrams from their correct straight positions but "incorrect" orthogonal positions, to their correct orthogonal positions. No big deal. The correction or workaround or improvement or whatever you wish to call it is too small to warrant the demonstrated upset, let alone the hysteria.
  3. I think many of Rob's assertions are incorrect, and manipulative, eg. that the Omni example "promises" symmetry in the orthogonal line instance, is simply not true. So I do not agree that Rob has ably demonstrated plenty of bugs. He has certainly demonstrated a few bugs, with unnecessary volume and repetition; and the bugs are too small to worry about for the majority of people; and we now know, they do not have enough votes to motivate Omni to proceed with.
  4. The sync bug is likewise, annoying, but simple to work around (just hit sync).

... Trying to work around [the auto-layout bugs] often exposes bugs in the Applescript support.
Yes, I have already agreed (in another of Rob's long threads) that there is an issue with the AppleScript Integration, which manifests as various bugs, it could definitely use some work. But I disagree that, as Rob posits, there is some "architectural flaw". And again there, Rob has some expectations which are not reasonable, and makes accusations which are not technically accurate, which diminishes Rob, not Omni or OmniGraffle.

And again, we have to accept, that Omni, being the size it is, has administrators who have a short term view, and who have prioritised their total work, the way they have declared, and there is simply not enough votes to deal with the AppleScript Integration issue.


The pricing of $50/$100 was most appropriate; the new pricing of $100/$200 places the s/w in the professional category, which means professional people use it, and they understandably have the normal expectations of professional s/w. There is a case (Australia & NZ, and some EU countries, where we have a very simple and direct Trade Practices Act) where we can demand errors to be corrected, unless the initial pre-sales documentation clearly identified that errors will not be corrected (which is the net effect of lumping errors and enhancements together and prioritising the lot). But hey, it is an American company, subject to American law, with the usual "no guarantee; condition is as is" declarations.

The notion that they are very interested in customer feedback is clearly false. No amount of opinion posted in writing has changed their model.


Last edited by DerekAsirvadem; 2012-03-19 at 12:48 AM.. Reason: Improved clarity.

I owe you an apology. The "not enough votes" part is quite correct. I now appreciate your frustration with Omni. Your tendency to emotional response and exaggeration is a bit more understandable.

But it won't help, swim within the lanes.

There is no piety in the organisation, the management is Godless. It is not hegemony, it is a small company with autocratic administrators who have a short-term view, in charge. I suggest your upset is due to your being technically professional (especially if you rein the rants in, to the technically accurate), and having professional expectations, which are common, and the large gap between that and what Omni have declared they do, which is not common; and therefore, you need to gnash your teeth and accept their commercial statement.

Also accept that, despite their declarations to the contrary and soliciting feedback, they have no intention of changing their model.

(I do not retract the rest of my post.)

Hi Derek :-)

All boils down to resource constraints, I think.

(Incidentally, I seem to remember some careful elaborations of why you were sticking with OG4 - perhaps you have upgraded since then, but if not, do think about it - the product can only progress if we go on paying the rent ...)
Originally Posted by RobTrew View Post
Hi Derek :-)

All boils down to resource constraints, I think.
I expect you are trying to explain Omni's position. I disagree. "Resource constraints", "over-stretched", etc, are the excuses of small companies. I am of the old school. Large companies with a long-term plan do not make change their long-term resourcing; they do not remove their QA; they fix all bugs, period.

(Incidentally, I seem to remember some careful elaborations of why you were sticking with OG4 - perhaps you have upgraded since then, but if not, do think about it - the product can only progress if we go on paying the rent ...)
That was years ago. I still have OG4 on the laptop. But since then, I have OG5 on two desktops. And the laptop is about to be upgraded.

No problem with paying the rent. But as of today, for reasons stated above, I have a serious problem with $200 for a s/w product where they do not fix bugs. I have recommended OG to many organisations without conditions, but from now on, since I work in Australia/NZ, I am obliged to clearly identify that they do not fix bugs as expected of professional s/w.

巧妇难为无米之炊 ...'

('even the most resourceful householder is hard pressed to assemble a meal without ingredients - or perhaps, 'where there's a will there's not always a way')

Something about their business model doesn't quite seem to be working ... too big a product range ? too strong an emotional attachment to small scale ? recruitment simply dampened by Seattle rain ? or by some phobia of tele-workers ?

Who knows ? In any case, some impression of a bottle-neck in the spreadsheets or in recruitment ...

Whatever it is, their pious invocations of the ballot box (in the face of long-neglected bugs) would not be so frustrating if their products didn't also have some real elements of technical excellence, and I wish them all well ...
Something about their business model doesn't quite seem to be working ... too big a product range ? too strong an emotional attachment to small scale ? recruitment simply dampened by Seattle rain ? or by some phobia of tele-workers ?
I've worked for companies large and small; start-ups that made it and ones that didn't. Omni come across more and more as a small company that has not broken the ceiling into the big time. I think you're right, unconscious attachment to small scale, without realising that their business model is severely hindering any chance of breaking the ceiling, and very attached to their way of doing things (including the business model that limits them so).

Product range too large. Yes definitely. But more to the point, they have a fixed number of developers, and instead of expanding that when they needed to (eg. implement the iPad suite), they've just spread that original number around, and therefore each product now has less developers. So we do not get the level of service that we used to get in the good old days.

Also, they have beefed up their product managers. In the old days, we had only developers responding on this forum, which meant the responses were top notch (because the Omni developers, unlike most companies these days, are top notch). But of course if developers spend their time on the forum, their development suffers. So they brought in various product managers, and some funny rules and attitudes. Now the technical quality of the responses is bad, but the politeness is top notch.

Omni has this idea that anyone should be able to pick up a product and use it effectively. That of course is not original, the beast himself started that in 1984. And it has destroyed the IT industry. Now we have florists and unemployed chefs developing "software", and erecting "architectural diagrams", all on the basis of some 3-week MS "certification" course. Personally I think the notion is na´ve, and geared very much to the short term, but consistent with the beast and the expectations of the younger crowd. In the context of their business model, it is geared to short term sales, single sales. Try before you buy, you too can draw amazing diagrams with this tool.

Uh huh. It's a drawing tool, not an artificial intelligence app.
I don't know if I am getting my point across, and it is an important one, so let me illustrate using a real world example. True story. I know a certain person who is an acknowledged master with databases, he has Data Models posted in many websites, answering technical questions. One day, the editor of an online database magazine approached him with "we want you to write an article, on how to draw beautiful Data Models, like the ones you have on that website".

Uh huh. The beast has a short attention span, 1500 words max. Not possible. He refused.

They really wanted the article, so after substantial back and forth, getting the editor to appreciate the science of Data Modelling; the science that went into the erection of the drawing; and the science that went into the stencil, they agreed to a series of seven articles, with a pre-arranged and agreed structure.

Now the possibility of their readership being able to draw beautiful Data Models, is actually realistic.
Sure, anyone can pick OmniGraffle up and use it, because it is intuitive, and that is because they have complied with Mac standards faithfully. (Many, many questions on the forum are really questions about how to use a Mac, not about how to use OG.) But the knowledge picked up in such an exercise is a superficial knowledge of the product, not what I would call a reasonable test drive. And if they end up buying, they then go about creating all sorts of complex drawings entirely the wrong way: without structure and standards; without templates and styles; using layers for all the wrong reasons; creating much more work than they need to; etc. The forums are full of people doing all the wrong things, and then wanting the product to fix their problem.

Second, the more they know about visual art; graphic design; cognitive science; the standards in their particular field; etc, the better their drawing will be, regardless of the the tool they use. A good tool will simply make it easier than a bad tool.

My approach is, sure, try the product out, but read the manual and go through the tutorials, before drawing anything complex. One saves a lot of time; avoids bad habits and incorrect practices; forms some structure; and builds an approach. Then one truly has a chance of drawing great diagrams. I was cautioned years ago by Omni that that was not the approach they wanted to foster on the forums, RTFM is a not allowed. But that is the more appropriate approach for long term sales, and sales to large companies that would buy hundreds of copies.

What I am saying is, their approach is consistent with their business model, and with short term sales, to single users; it is inconsistent with long term sales and corporate (hundreds of copies) sales. So they are stuck in the small end of town.

The voting on bugs plus new features plus enhancements (to existing features) all lumped together, is "fine" for a small company (limited resources) and a short term vision. It is antithetical to either a large software company or the long term approach. And a source of frustration to those of us who expect Omni to respond like a large company with a long term vision.

As for the product management and marketing, they should move away from selling pretty drawings (with OG being the enabler), and instead sell a brilliant drawing tool (with pretty drawing being the result). It appears they have not realised what they actually have. OmniGraffle rocks, but no one in the corporate world has heard the tune. Once they realise the market exists, they will start marketing in the right circles, and then the business model will change.


Last edited by DerekAsirvadem; 2012-03-21 at 04:51 AM.. Reason: Added clarity; corrected typos.
Thanks for the answer Derek.

However, this will not fix the bug we have.

I need to connect the line to another line, not to the object. Using straight lines will create a big confusion.

We are creating a simpler way of flow charting so using out-of-the-norm flowcharting convention is something that has to work here.

By the way, I have the pro version and have the bug.

Here's a Omnigraffle file and two screenshots of a line not connected with anything and it has a bug.

Hope Omnigraffle can fix it.
Attached Files
File Type: pdf sshot01.pdf (34.3 KB, 418 views)
File Type: pdf sshot02.pdf (36.5 KB, 410 views)
File Type: zip (30.5 KB, 394 views)
Does switching to a slightly different zoom level restore visibility to the partially vanished links ?

( This bug, and the family of OmniGraffle screen display bugs to which it is related, seem to be rooted in a misfired attempt to improve screen speed by skipping certain kinds of 'unnecessary' display updates ...

Typically, the zoom level in which the error first occurred remains corrupted (data or recent edits partially invisible or misrepresented), but other zoom levels may fully display the data in its correct and current editing state).

[A conversation with support some weeks ago suggested that the chances of Omni actually fixing it before another major version is developed seem to be rather slim.]
If some variant of a script of this kind proves palliative, it might be worth assigning it to a keystroke ...

tell application id "OGfl"
	tell front window to set zoom to (zoom * 0.999)
end tell

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Similar Threads
Thread Thread Starter Forum Replies Last Post
Tables with Horizontal Lines (but not Vertical Lines)? lanceosborne OmniGraffle General 6 2010-11-11 02:40 AM
Favorites disappearing avilewis OmniWeb Bug Reports 21 2010-03-12 06:27 PM
Disappearing projects Caroline Wilson OmniFocus 1 for Mac 1 2009-11-02 02:29 AM
disappearing indents Suzannah OmniFocus 1 for Mac 4 2007-12-12 07:41 PM
Status bar disappearing act OmniWeb Bug Reports 3 2007-01-23 11:37 AM

All times are GMT -8. The time now is 03:44 PM.

Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2022, vBulletin Solutions, Inc.