Welcome to the Onshape forum! Ask questions and join in the discussions about everything Onshape.
First time visiting? Here are some places to start:- Looking for a certain topic? Check out the categories filter or use Search (upper right).
- Need support? Ask a question to our Community Support category.
- Please submit support tickets for bugs but you can request improvements in the Product Feedback category.
- Be respectful, on topic and if you see a problem, Flag it.
If you would like to contact our Community Manager personally, feel free to send a private message or an email.
Standalone application for Mac OS X, Microsoft Windows, and Linux.
Michel_Storms
Member Posts: 11 ✭✭
in General
Hi,
is anyone else with me on the thought that Onshape needs a standalone application? This could be done using a self-branded browser that is optimised for using Onshape. Why do I think of this? Because although it is nice to just go to www.onshape.com and start designing, people who need and rely on it in their daily workflow don't want to be dependent on third-party browsers, but want a reliable application. It is just not right to tell people that Onshape does not work with Internet Explorer, Microsoft Edge, and does not run optimal in Safari. It is not right to send people to browsers like Google Chrome and Mozilla Firefox. It is awesome that it can run in these browsers, but that should not be the primary way of using it. You should be able to say 'We have an application that is available on the Microsoft Store and App Store, but Onshape also runs in your browser. For that, we do recommend Google Chrome or Mozilla Firefox on all platforms.'... You can not tell people which browser to use, and that they will be dependent on it for their daily workflow. Please read this post thoroughly, it is based on my experience in introducing Onshape to professional people I know, and a correct stand on this matter will make or break the future of Onshape.
is anyone else with me on the thought that Onshape needs a standalone application? This could be done using a self-branded browser that is optimised for using Onshape. Why do I think of this? Because although it is nice to just go to www.onshape.com and start designing, people who need and rely on it in their daily workflow don't want to be dependent on third-party browsers, but want a reliable application. It is just not right to tell people that Onshape does not work with Internet Explorer, Microsoft Edge, and does not run optimal in Safari. It is not right to send people to browsers like Google Chrome and Mozilla Firefox. It is awesome that it can run in these browsers, but that should not be the primary way of using it. You should be able to say 'We have an application that is available on the Microsoft Store and App Store, but Onshape also runs in your browser. For that, we do recommend Google Chrome or Mozilla Firefox on all platforms.'... You can not tell people which browser to use, and that they will be dependent on it for their daily workflow. Please read this post thoroughly, it is based on my experience in introducing Onshape to professional people I know, and a correct stand on this matter will make or break the future of Onshape.
0
Comments
But I also agree that for the moment such an endeavour would be several bridges too far (spreading the development effort way too thin)
I don't disagree that Onshape should always run on at least one mass-market browser on every major platform - that's a given.
So the dedicated browser (or standalone app, if that's what it takes - but that would be logistically difficult with upgrades) seems to me an "AND" proposition, not an "INSTEAD" one.
And, if it is ever a necessary solution, it will be to a "nice to have" problem, because it implies major success.
First succeed ... then augment, redouble, and transcend ...
-1 if it's just one more software requiring updates without noticeable advantages (other than not being firefox or chrome)
The other issue of having local installed Apps, and not necessarily browsers, is how easy things should be to share a design with anyone. Onshape will push to be more "YouTube - like" as far as sharing design information easily. I think 3rd party browsers are superior for that. I would hate for someone that I share a design with to get a message... "install this ----- first, before you can look at it". That would not bode well with IT departments either. Then someone is on the phone with another manager trying to get something installed. An hour later, lots of questions, a new installation, and finally you are looking at that design. No, it's too many hoops for the casual user.
See cake, want cake, eat cake. Still have cake,
I find I am forever "holding the hand" of different browsers trying to diagnose or nurse them through different performance collapse issues. Am exhaustive list would be .. exhausting, but to take a couple of examples from Chrome on a variety of platforms, I've struck several crippling extensions compatibility issues, and there can be problems (not least forensic ones, but possibly also performance ones) arising from the way a separate process gets launched for each session, each tab and each extension, .
I don't like the potential vulnerability to trivial third party browser updates "breaking" key functionality in Onshape. It's already slightly worrying that upgrades within Onshape get pushed regardless of whether we're hours from a key deadline of a major project, but that gets compounded if the app runs on a free piece of software from a supplier who is accordingly not beholden to us, and who is stretched in seventeen directions trying to be all things to all people.
I chafe at the unavailability of certain universal interface elements (like Control-drag to copy) because certain browsers hijack certain keystrokes, and this seems to me a visible instance of something of more concern, a small tail gets to wag a large dog.
On the proliferation and complexity front, I don't personally think the dedicated browser would need to be multiplatform or multi OS capable.
Serious users of Solidworks et al routinely submit to a tyranny on the question of choice, so the increased number of options for Onshape to support could be as small as one.
That said, I disagree with having a standalone desktop application or mobile app for onShape. Here's why. The beauty of onShape is that you can access the same application on any tech platform that has access to the internet and a (certain) web browser. The fact that it specifies the use of a certain brand of browser is, IMHO, just a minor detail. Having it as a online web-based app would mean the developers never needs to maintain the software for multiple platforms. Imagine if they released a standalone desktop application for Mac OSx and Windows. That would mean they would have to maintain two software packages every time they release an update. Same thing for iOS and Android. You may realize that you can get code libraries and compilers that allows you to write code and compile it to BOTH platforms. But those who have experience doing this knows that its never that simple. Lets take Microsoft Office for an example. Why don't they synchronize their Mac OSx and WIndows releases for Office?
Another advantage of having a web-only application is that onShape never has to give anyone their software. That makes it hard to pirate. Also onShape is in charge of the updates and they can synchronize those updates for all their users. They no longer have to support users who insist on using the same software for 12 years and expect it to be compatible with new versions. Remember Microsoft's Compatibility Packs? All of this adds up to big cost savings that they can use to push the product forward rather than spending time and precious developer resources on things like preventing piracy, building compatibility packs, making multiple versions for multiple platforms not to mention the testing that would require. THey don't have to worry about software patches, the software exists on their own servers. There is a steadier stream of income when users pay per month... and that benefits the users too as they don't have to spend $30,000 every few years on a major software update.
The benefit for the end user and company is that ... all the IT personnel have to supply is a computer with a web browser and internet connection. They don't have to worry about pushing out software updates to all their client computers, sychronizing updates, storing / backing up / securing the data. Web Apps are an IT departments dream come true.
I use onShape on a $250 Acer Chromebook, 2GB RAM and i3 processor. It works flawlessly. Very responsive because of the extremely light operating system (ChromeOS).
So all things considered, I'm putting my support with the current direction of onShape.
Andrew, if its a Windows or Mac computer, it could be because Windows and OSx is a pretty heavy operating system. You have this HD operating system and all you're doing is running a web browser. Like driving the Ferrari to Safeway. My work computer is an 8GB i7 with a 1GB video card. I actually don't notice any difference between using it and my 2GB i3 Chromebook. I actually like to ue my desktop to watch the instructional videos for onShape because it has better sound and a big screen, and I do my onShape on my Chromebook. onShape on my cheap Chromebook is incredibly responsive.
We also have a 4 year old set of Samsung XE303 Chromebooks. Thats a Terga processor (basically a Galaxy Note processor) with 1GB RAM and onShape runs just fine on those too.
By an extension of your argument, is it not theoretically possible that off-the-shelf browsers are also (in your words) "pretty heavy" to be running an app like Onshape?
You previous post appears entirely targeted at a stand-alone app. I'm curious how you would feel (at some future date, when Onshape has gone viral and is fully capable) about my suggested alternative: an *optional*, dedicated-to-Onshape, stripped down and optimised, lightweight and highly reliable browser? One which connected only with Onshape's servers, and would consequently not need to be heavily defended against malware or hijacks?
@stormychel I have had similar experiences in more "technologically stagnant" manufacturing environments with a workforce that just knows the browser that's in front of them. I would expect awareness of better browsers to begin to improve as we roll forward. That doesn't help the today storyline, though. It would take some educating on your part and either admin rights to install a better browser, or a good reason to have your "technologically stagnant" IT department to do it. I wouldn't vouch for Edge at the moment. I have run it natively and run into issues. As I'm aware it's missing things in order to be fit for a more proper use with Onshape at the moment. My preference is Firefox over Chrome on Windows, just because the gain on the zoom (scroll wheel) is higher and quicker to zoom in and out of where I want to be. I'm generally not working with more massive models yet, but wonder if Chrome might outcompete Firefox on speed if I were to check.
Failure of a dimension to update when edited.
https://forum.onshape.com/discussion/comment/12034#Comment_12034
To rectify this is a simple matter of reloading the browser tab, but this can be time-consuming for complex models or slow connections, and disruptive to trains of thought even when it happens quickly. And it's a barrier to new users who don't know what the problem is or how to fix it.
If it's not the browser's fault, then Onshape need to fix it, but it has been a known, ticketed problem for many months.
If it is the browser's fault, then there is no way for Onshape to fix it. And that class of problem seems to me of serious concern, for anything more than hobby use.
In Chrome, I've struck several very perplexing problems which arose from using zoom levels other than 100%.
The worst of these involved feature list items disappearing from the model instantaneously, without warning or visual cues, simply by RMB on the item in the list. AFAIK it's still a live issue.
Imagine the damage (say) a junior collaborator could do during a prolonged shared session, if the disappearances (and consequent damage to dependent portions of the model) were interleaved with real and consequential edits by the other participant, working further up the tree, and consequently unaware of the mayhem happening out of sight. When I think of some models I've worked on, with hundreds of modelling hours hanging on a slim thread, the mere suggestion makes my blood run cold.
Sure, the junior should hold their hand up at the first sign of problems, but who among us has never kept furtively and frantically trying to dig themselves out ...
when they should have stopped long ago and sought help
(the first stage of which will inevitably involve said junior digging themselves out of a steaming pile of trouble and strife)?
Intricate interleaving (it seems to me) means that restore from History will inevitably lose all the work done since the problem began. I cannot see a way a "selective undo" could ever be implemented, given the degree of mutual feature dependency inherent to a parametric, history-based solid modeller.
Admittedly I can't be certain that this is a compatibility issue between the browser and Onshape (I guess it's conceivable it could all be Onshape's fault) but if it is about compatibility, it's not something Onshape alone could fix without writing a dedicated browser.
It would only take each compatible browser on a given platform to throw up just one such potentially showstopping problem, and Onshape would be hung out in the wind.
Didn't work but I really didn't expect it to. Just curious how far I can go with a $50 computer. Looks like some Pi 2 fans, here is my Pi 2 setup. Just basically playing with it.
Thats a neat idea. Chrome is built from the Webkit library which I believe was initially an Apple initiative before it became open source. onShape likes Chrome. So following that logic if onShape were to release a browser to use with their app it would be built from a Webkit library.
To be blunt, if I were in charge at onShape, I wouldn't do that. Then they would have to divert resources to developing that browser and away from CAD. They would have to develop one for each OS, Linux, MACOS and Windows. Then you run into that problem again of having to maintain and update three diffrent software versions for each OS, people not updating their onShape browser etc..
Given the cost vs benefit, not sure if its worth it to do that. Just my $0.02
Andrew, this is somewhat unrelated, but reading your post reminds me of a recent update to Google Apps.
I hope you are not surprised to learn that I'm a pretty big user of the entire Google Apps platform including some development in Google Apps Script. There was a fairly recent update to Google Drawings. They are now fully collaborative. They were not before, if you were editing a drawing and the browser or computer shuts down, you loose your edits. But not anymore and it now allows for realtime collaboration.
Doesn't onShape have some kind of a bookmark feature where it can revert back to a certain state of your work no matter how many edits were ahead of it?
A model is not like a 2D drawing, where a problem in one corner of one sheet cannot bugger up a view in another sheet.
I'll be documented my struggles some more. Given that others have run similar resource intensive WebGL games on the Pi there is a possibility this works though admittedly it's looking like many other solutions will be more efficient in the short term.
Link to my struggles here:
http://www.3dnpd.com/2016/08/22/the-35-cad-computer-onshape-on-a-raspberry-pi/