ttd

13 ms·

121.
by ttd 9y ago
Absolutely -- it's an important feature for me as well. I have a development branch already with initial support for "sticky" lines with this behavior, but it did not make it into this release.
122.
by ttd 9y ago
At this time, there's no native import for GraphML, although this is something I do want to support eventually, including other graph specification languages like DOT. This is not implemented yet, however.
123.
by ttd 9y ago
Radial arrays are a planned feature. I'm still undecided on how the UI will work (if it will be similar or not to the existing duplicate on a grid tool), which is why it didn't quite make it into this release. You can define your
124.
by ttd 9y ago
This is a great point, I will definitely add a blurb in that section. The trial version is not time-limited, but it is limited in two ways: each drawing can have a maximum of 30 distinct objects, and exported images/PDFs have a small w
125.
by ttd 9y ago
The slideshow looks like a great reference, and I will absolutely be revisiting your Lua sandboxing repo to learn more about your approach. Thanks!
126.
by ttd 9y ago
I haven't done that experiment yet, but it's an excellent idea.
127.
by ttd 9y ago
I'm doing some small amount of sandboxing for the program-mode drawings, but right now there's nothing preventing an infinite loop of primitive creation from exhausting memory and eventually crashing the application. Switching bet
128.
by ttd 9y ago
Instantly in this case means several things to me (person who wrote the copy). 1) Very quickly create precise diagrams, meaning diagrams where you know things are perfectly snapped, aligned, etc. 2) Very quickly see the changes you make to
129.
by ttd 9y ago
On the roadmap is a "shape search" similar to what Visio has. Is that the sort of templating you're thinking of?
130.
by ttd 9y ago
Rats... do you know the resolution of your phone screen by chance? Good luck with the thesis!
131.
by ttd 9y ago
Yes, PDF (and PNG) export are automatically cropped by default. You can export an arbitrary region of the drawing as well. You can also just export the objects that are currently selected, which is convenient for doing repeated exports of d
132.
by ttd 9y ago
Thanks for the feedback! There's nothing in Vexlio's software stack that would prevent a Linux version (thanks to the Mono project), but it would take a non-negligible time investment to make it happen. I'll have to see if th
133.
by ttd 9y ago
Plugins are definitely on the roadmap. Much of the complexity there is just exposing the API, which is already done for program mode. Your idea of generating diagrams from web request data is great! I'll definitely keep that one in min
134.
by ttd 9y ago
Thanks for the compliment! Vexlio itself is written in C# (the Mac version in development uses Xamarin.Mac). I'm using the open-source NLua library to provide Lua hooks to call into the application.
135.
by ttd 9y ago
In your scenario, the way to gain access to the objects you created in normal mode (e.g. to modify their properties) is by using the `Layer:getObject()` method to retrieve objects by their index. You can see an example in the "Modifyin
136.
by ttd 9y ago
I'm hopeful that the price does seem competitive, but not too low. I'd say the Visio market is slightly different, although you're right there is probably some overlap. Thanks for checking it out, and I hope to have a Mac ver
137.
by ttd 9y ago
Hmm, I'm seeing some slightly slower-than-usual response times but it still seems up to me. Hopefully the issue clears up for you.
138.
by ttd 9y ago
Thanks! The scripting language is Lua 5.2.
139.
by ttd 9y ago
Hey HN, Creator here. I thought this would be of interest to the HN crowd for things like the LaTeX equation integration and Lua live-edit mode. Happy to answer any questions!
140.
Show HN: Vexlio – Create precise, beautiful diagrams(vexlio.com)
309 points by ttd 9y ago|130 comments
141.
by ttd 9y ago
Optimizing memory and cache use has been a primary focus of compiler research for a while now. I'm don't understand why you say partial evaluation w.r.t. optimizing memory use is a neglected area though: I think partial evaluation
142.
by ttd 9y ago
> Applications today are all HTML websites. Perhaps I misunderstood your point, but taking it literally: this claim is just plainly untrue, and reflects a very narrow-minded view of software in general.
143.
Ask HN: Worth filing provisional patent for micro-ISV product?
1 points by ttd 9y ago|1 comments
144.
by ttd 9y ago
I would guess it's based on methods from Wi-Vi [0], so the ground-breaking was done in 2013 in academia. [0] http://people.csail.mit.edu/fadel/papers/wivi-paper.pdf SIGCOMM'13
145.
by ttd 10y ago
I'm not an expert by any means, but I think you can manually gzip your content and add rewrite rules to Apache (don't know about other servers) to use the .gz versions. I think this is the only way for compressed content to be cac
146.
by ttd 10y ago
I'm curious about this. If you have a purely static site, what would be the major bottleneck for a front-page-reaching site hosted on a raspberry pi? I'm thinking things like compression would eat up a lot of its CPU, but you seem
147.
by ttd 10y ago
Any thoughts on WPF in comparison?
148.
by ttd 10y ago
I don't really agree with the "playing the system" sentiment. It isn't as if we took the same paper 10 times and just submitted it until it stuck somewhere. Each time we got valuable, actionable feedback, which we then a
149.
by ttd 10y ago
Not to my knowledge. Gaining familiarity with the different venues in your subfield is an important part of the graduate (and beyond) education.
150.
by ttd 10y ago
In my experience, papers often take more than just two submissions to be published. It's often a question of finding a good community fit. One paper I coauthored took 8 years and probably 10 submissions before it was finally accepted.

More ›