Hacker Newsnew | past | comments | ask | show | jobs | submit | throwawayfinal's commentslogin

Depends on the state. The majority of states either have laws relating to disclosing known STDs or laws specifically about disclosing known HIV status, or both.


If you're building a native python application, there's no reason to adopt yet another technology.

A native python app in "Docker/Containers" is going to look like a dockerfile that includes a copy of the app and runs three commands. Either you build that on the server (what's the point) or use a registry (additional complexity for little benefit).


displaying language is full of corner cases. imagine having to render both english, japanese, and arabic in the same code paths, and inline with each other

text display code is not so simple as it seems, and don't fall into the trap of "how hard could it really be"


I'm not saying it isn't hard - I acknowledge that getting it right is a huge task. But the cost of failure should be rendering incorrectly, not crashing or memory corruption.


There was a similar bug about a year ago as well. I'm surprised more fuzzing, static analysis, and general testing hasn't gone into this area since there have been issues found there already. Maybe they have done this extra work and just didn't catch this one. It's hard to know. Still feels like the way it fails is unacceptable.


"Especially important for finite supplies like Bitcoin"

The law doesn't concern itself with the technical implementation of bitcoin. In addition, trackable and "tainted" money is dangerous for people to accept; bitcoin just makes it real easy to identify that.


"The law doesn't concern itself with the technical implementation of bitcoin"

That's an astounding statement to make without citation or qualification. In general, the law (at least in the form of U.S. appellate courts and congressional bodies) is very good at including public-policy consequences in its decisionmaking process.


maybe future laws and policy will take into account the more-or-less finite supply of bitcoin. maybe.

my point is that there is no guarantee that bitcoin will receive favorable decisions. it's hard for the lay person to understand (why not just print more?) and if ever declared a currency-non-grata could actively be attacked in this way.

i should add that i have engaged in btc currency speculation and had a positive result. the long term value/health of btc is of great benefit to me. but, it would be foolish for me to assume that because it would good for it to be so that it will be so.


That much I can agree with. The law eventually gets things right, but it can take a while, and it might harm individuals and individual technologies along the way.


If Google is using GCP internally, then their infrastructure is as you said.

The GCP "best practices" sections in the docs* have been top tier in my experience and you can learn a lot, even if you're using AWS or Azure.

* my one complaint is that searching for the docs almost always ends you up on marketing pages, which you then have to find the docs link on.


People don't love XMPP for a lot of reasons. Some of those reasons apply to IoT. Some reasons, like having to maintain state, are IoT specific.

Besides that, bandwidth and code complexity is a very real concern. MQTT, a lightweight message passing protocol, has become the de facto default.


So true. We switched to MQTT (abandoning websockets and other protocols) to save (a lot of) network bandwidth.


Do you mean you were using MQTT over websockets before, and you've since moved to pure MQTT (still over TCP), without websockets in between? Are you sure websockets were the cause of your network bandwidth issues?


I'm on a different team, AFAIK we where using simple JSON payloads on websocket connections. Dropping websockets and switching to MQTT saved a lot of bandwidth. I can ask for precise numbers or details if you wish.


I'm interested.


Please expand


As a developer, Electron's page shows their prowess in writing good front end code, which is really important. As someone who understands the basic concept of Electron, I want a product that demonstrates those competencies.


I get what you're saying, but "writes good web front end code" isn't on my list when I'm looking for a good Desktop development framework.

I think stuff like this is pretty tribal, which is OK, but we should confuse it with decisions made on technical or experience merits.


We shouldn't, but we will.

We're still savanna primates, and silly little things like technical merits aren't going to get in the way of sexy shiny. Projects that understand this fact will attract developers and flourish, and those that don't will languish in forgotten obscurity.


When the "good Desktop development framework" is primarily in javascript, it's a good signal.


There are a lot of counter-examples:

  - apache

  - lighttpd

  - v8

  - PostgreSQL

  - Linux

  - Go

  - ZeroMQ

  - any imaging library (libpng, libjpeg)

  - any compression library (zlib, xz, bzip2)

  - urxvt

  - vim

  - emacs

  - tmux
Basically none of these have good websites. When I'm looking for a CSS framework or like, UI tricks like scriptaculous then I think it's a good signal, but past that I really think there's no link here. Really "has a good website" means "someone who knows how to build good websites built one for this project".


I would expect that apache and lighttpd are hosted in their respective projects; linux is used to run linux.com servers, etc.


Sure because those are examples of what you can do with the tech. You can't really make a website with Qt, so using... PyQt's website as an example for evaluation doesn't make a ton of sense.


Is it really that good though, or is it just that a lot of design effort (and dollars) have been put into it? Just based on the quality of electron based applications I have used, I feel like this basically amounts to spending a lot of time polishing a turd. Why compliment or be distracted by how shiny it is?


More than once, cutesy comments of frustration have led me to inspect something for side effects and avoid them.


Could it not be done through conventional ways like TODO or FIXME?


Sometimes there's not something to fix or do, for example, in pthread-using code. "Just be really fucking careful or we'll all cry a lot."


This begins to fall apart in major cities without OSM volunteers ingesting city roadwork feeds. Of course, something like tomtom wouldn't represent that ideally either without a radio data or cell service...


Waze used to (still does? I don't know as I don't use Waze any more) display dots on paths and as you drove it scored you and also used the data to confirm maps. Is there anything like this in the OSM world?


Drivers in the field can record GPS tracks, and upload/share them through the OSM website. These can then be used either to directly create that road, or by remote mappers to help align satellite imagery, and then remote map visible roads in the area.


Not really. But OSM is a wiki. So there is no central body to "confirm" changes. If you want, you can just edit things as you go.

You can even edit OSM on your phone (e.g. with Vespucci), so you could drive along a road, think "this is totally wrong on OSM", pull over, get out your phone and correct it.


OSM volunteers are pretty crazy, though. A friend of mine spent his time doing linework for North Korean roads. It's a data sparse region for us capitalist pigs. Anywhere, there might be good data where you least expect expect it.


Learn word. It's still best in class.

Or LibreOffice. Depending on the type of manuscript, Dryad or celtx.

Of course, I don't know your workflow.


Yikes, Word on macOS is certainly nowhere near best in class. It's a completely buggy heap of trash, as soon as numbered styles, images, or tables of contents become important. On the plus side, Word doesn't crash as much as Excel on macOS does for me. Pages, Numbers, and Keynote are far better on macOS than their respective MS apps in my experience.

If I had to use Word for anything except the most simple documents, I'd probably run it in a VM.


Hopefully, that will change in 16.9.0 with Office for Mac sharing a codebase with Office for Windows:

https://arstechnica.com/gadgets/2018/01/office-for-mac-final...


Been beta testing this for quite some time, if anything it’s getting worse. The performance of outlook and word is abysmal, random crashes, high CPU usage under no activity and the list goes on.


Ah, so it will be crashing on both platforms! Excellent.


Is this with Office 2016? Almost sounds like the 2011 version. I'd say it's a resource hog but it doesn't seem particularly buggy.


Better yet, don't upgrade. MacOS upgrades seem to just be getting worse and worse.


I tried word on macos. I recall there were some issues which made it harder. Though, I would like to use it, as Microsoft will support it until the end of time.

I may simply switch to a markdown template. I used to use Pages for book publishing and copying to html, but the books have become less important.


sidenote: Pages indeed creates some of the best html of any Wysiwyg I've looked at.


It's astonishingly good. It's why I'm thinking of markdown as a replacement: I know Markdown converts to html. Other editors - not so much.

It was a lucky accident for me that Pages worked for converting my books to html.


Word is truly dreadful, not just to use but the code it generates in each document is a mess, it’s no wonder there’s so many formatting problems in docx files.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: