Hacker News new | past | comments | ask | show | jobs | submit login
Adium Redesigned (nouincolor.com)
168 points by pchm on May 1, 2013 | hide | past | web | favorite | 103 comments

The drag-and-drop file sharing, with inline preview, is pretty amazing. I wish Adium had that.

I really dislike the rest of the UI though. One very important thing about Adium is how compact it is. Even on relatively small laptop screens you can just leave it in a corner, which makes it easy to hold conversations without having to break context.

The conversation list, and the conversation views themselves, feel very much like the much-maligned "gigantor" Skype update from a while back. Everything feels much too big, much too spaced out, for a pure utility app. IM isn't an "experience" app, it's about as utilitarian as computer use gets.

The "smart command line" thing is very cool though.

The compactness of Adium is key. I always set it up to have my friend list as a translucent border free floating window to where it is unobtrusively sticking to the top right hand corner of my screen. I enjoy only seeing tabs of people I am currently chatting with; the list to the left on these screens is too bulky.

> The conversation list, and the conversation views themselves, feel very much like the much-maligned "gigantor" Skype update from a while back.

mixed with facebook message since you can view a preview of the messages as well. I think this is how Skype should have done it.

Exactly, this doesn't look like Adium, it looks like a flat version if Apple Messages.

Wow, why is there such an unbelievable amount of negativity and trolling about this project? I applaud what Oskar and Dawid are trying to do. Adium is a great project, but its starting to feel a little stale and could use some people experimenting with different UI.

I hope Dawid and Oskar find somebody to help them with this experiment and I hope encouragement and constructive feedback start float to the top of this thread. Lets help these guys get help and see what happens!

It's no surprise, it's classic middlebrow dismissal that plagues most submissions these days, particularly design based ones. Thankfully 'flat' or 'skeuomorphic' aren't mentioned in the submission to make the comments a real warzone.

I'm not saying that it's impossible to receive some good design advice on HN as there are certainly some experienced design-based folks dotted around, but on the whole it's a bit like asking for database optimization advice on Dribbble.

Counterpoint: nothing is more middlebrow than thinking that something needs updating just because it's old or not "flashy".

If you read the article, you'd see it wasn't a post seeking advice. It was a post where a mockup was made, some pie-in-the-sky new features were recommended, and it concluded with the poster saying someone should do this.

This wasn't an invitation to join their project or even any mention that they would spend any time on it themselves. Hell, they didn't even include the source graphics to start anyone off.

So...what's the problem?

Probably for the same reason developers are negative about people coming up and saying, "Hey, I have a great idea for an app/website/..."

They want to fork the project and build a new UI? Cool. Go do it. If people like their changes they'll start using their version of Adium, and other people will contribute to their project.

>"Hey, I have a great idea for an app/website/..."

It's not just an idea. It's a considered UI for the community to critique and possibly implement. An open source project thrives on multiple people to help out. If this mockup was so simple and obvious, it would have existed already.

Unless I can print this out and rub it on my screen to create the code and associated elements needed to make this work, I don't see how this is anything other than a mockup...which is an idea.

That's like saying floor plans or a logo is just an idea.

If I draw up a floor plan for Cowboy Stadium and post it on the Internet that I think this is how the new Vikings Stadium should look, I believe that would be called an idea.

The posted concept isn't just a copy--there have been decisions made based upon functionality.

To follow the analogy, one could use the Cowboys stadium as a precedent and then adapt it to the climate of Minneapolis.

This isn't a discussion about the codebase because the code can't solve the questions being discussed. There isn't anything in the concept that is out of the realm of feasibility. Critiques happen at all phase of design and development, and you seem like someone who's uncomfortable with ideas that aren't finalized.

But if you have the contract for designing a new stadium, then it's actual work?

No, if I can't run it on my machine then it's just an idea. I didn't say it was a simple or obvious idea - I said it's just an idea.

Anybody can make something pretty in Photoshop. If he was serious about getting people to implement his mockup he should have spent a few days writing a quick and dirty implementation, with real code that could be improved upon later.

This sums up why most open source GUI's look like a widget library vomited on the screen, don't try getting involved unless you know how to code.

It's just a chat app. We all know that programming it is feasible because it has been done hundreds of times before. That part isn't interesting.

What's interesting is the interface, which is why we're discussing it. The fact that you can't run an application doesn't make the concept any more or less interesting to discuss and critique. The issues being discussed aren't solved with code.

Lol? Anybody can make something pretty in Photoshop? I dare you to try and make something similar.

You know that there's a profession called UI Designer?

Usually those people do not talent or can't execute. Oskar and Dawid have already executed a great design concept and are doing an amazing job evangelizing it in hopes of recruiting a developer to help.

If I had the time and knowledge of the Adium code base I'd be thrilled to collaborate with these two talented individuals. Hopefully there are developers out there with open enough minds, time, and passion to work with these two.

I look forward to seeing you, Oskar, and Dawid all leading the forked (or new) project soon. Please let us know when the repo is available and what you need help with!

I don't have a problem with this. I think it's great to imagine and mock up improvements to an app. But at the same time I don't have anything good to say about the result either.

The fact that something looks "early 2000s" is so far down the list of my priorities for an IM app that it's effectively a non-issue. I think a lot of the sentiment here is simply that the important issues aren't really being addressed, and implementing a radical change like this is likely to raise a whole slew of new ones.

I see more criticism toward Adium and Messages. Just because people don't agree with what was done here doesn't mean they are automatically trolls.

If Dawid and Oskar want to progress from mockup to making this a project, I wholeheartedly encourage them visit http://www.adiumxtras.com/ and learn how to reskin Adium. Their talents clearly show that if they put even a minimal amount of time into it, they would be able to accomplish most of the UI on their own.

It's not that the idea of redoing the UI is a bad one, but the execution reminds me too much of Skype. Skype went from a compact, easy-to-set-aside layout to an elephant on the desktop. It's ridiculously huge now (size-wise).

Can't agree enough. When I'm using Skype I almost have to stop everything else I'm doing thanks to it's massive window taking up a whole corner of my screen and constantly having to hide/unhide the messages pane.

Adium is fine as is, you can customise the hell out of it, it's tab implementation is solid. The only places it's lacking are more server issues, file transfer hasn't worked in years for me.

> its starting to feel a little stale

The concept of "staleness" encourages change for the sake of change, and I don't get it. Please ask non-designers if they want the look of their favourite software to be re-imagined.

The feedback to this is predictably similar to the transition from Skype 2.8 to Skype 5 (OS X), and I know people who still cling to their Skype 2.8 binary.

When something looks the same as it did years ago, while the entirety of the OS is being updated, not to mention design trends are changing, "stale" is a valid description. It's not "change for the sake of change" it's change for the sake of keeping up with the newest cool design trend.

Average consumers will choose a nicely designed app over one that looks dated, especially if they offer the same functionality.

But OS X does look the same as it did years ago, unlike Windows which went through five looks (2000, XP, Vista, 7, 8). It's not as if Adium has any design elements that could possibly age - it's a standard OS window with a customisable HTML message list.

Edit: I just realised that I have hidden the ugly message window toolbar. I'm not sure if it aged though - I already hated it on OS X 10.4 (when I disabled and then forgot about it).

"Adium Reskinned" is more appropriate. This is a mockup that some designer did that fused (the bad parts of) Messages with the background color on this blog and then posted it on HN.

Maybe the designers should learn how to skin Adium and do this? I fail to see why they can't?

There is a limit to what you can skin in Adium and some of those things you just can't change so even if you make the messages view look beautiful the tabs for each conversation are un-theme-able.

Nevertheless the UI skinning can be mostly done. I see no reason why the designers couldn't be doing this themselves.

I am kinda sick of titles that suggest someone has actually done something more than just create mockups, when really it's just some mockups.

We can all make some mockups in an afternoon. It's not really worth telling everybody about unless you've done something, IMHO.

Making mocks is doing something. I don't think it's healthy for programmers to discount the efforts of designers or vice-versa. Now, it's easy to make mocks that gloss over the real challenges or nasty corner cases, but it's also easy to write garbage code that will never be useful to anyone.

The fact of the matter is that there are a lot of people out there with lots of engineering ability and not much design ability who might jump at the chance of implementing something that's been thoroughly mocked and product-ed out.

I don't discount the work designers do, but making mockups is only the start of a designer's job, IMHO.

It's like a programmer saying "Job's done" because they made a DB schema.

Designers depend on programmers to take it from mockup to prototype. How can that be done if nobody knows about it?

You are saying that a mockup is basically nothing, that anyone can do it. If that's true, why do we need designers at all?

This is going to be an unpopular opinion on HN, but I would say the need for designers is hugely overblown.

Most of the time the only people I hear complaining about UI design are UI designers. Except for exceptionally bad cases, most people don't care too much what their applications look like.

And most people would rather have a working implementation with an ugly UI than a pretty picture/mockup that doesn't do anything.

IMHO it's the designer's job to make the UI. That means providing the programmer with templates, CSS, and UI related JavaScript.

If all the designer can do is mockups, then IMHO they are not doing their full job. Like I said in the other reply, it's like a programmer saying "I made the DB schema, what more do you want?"

Not 'some' mockups. One mockup. A single image.

Not even any explanation or justification of design choices. Just a sentence claiming it is 'self explanatory'.

And then there is the last sentence, which basically just means 'who wants to take my single concept image and do all of the remaining work?'

Man, that is an example of graceful degradation, just not what we usually understand this to mean.

- Bubbles. Seriously? The 2005's want their failed metaphors back. Thanks, the list-and-icons Adium presents are way more useful

- stamp-sized images to disturb the text flow. Great idea, please forget about it right now.

Adium is completely unsexy, and this is a good thing. I don't need a chat application to take over my whole screen, this is not Facebook Home.

Next up: Terminal.app reborn: output of Unix commands grouped by "conversation" - because is looks so much nicer. Icons for every shell command - much sexier looking.

Your Terminal.app reborn example sort of reminds me of https://github.com/unconed/TermKit

Thanks, I need a Whiskey now ;)

Everyone is not you. Some people like to have a chat application take up the whole screen. And chat bubbles are alive and well, for example in Apple's own Messages.app

I refuse to believe there is a single person on earth who actually likes the OS X Messages.app

I switched from Adium to Messages because I haven't found a way to turn off annoying dock animation of Adium when Internet is offline or when jabber server is down.

Later I found that Messages' proprietary protocol that synchronizes chats on Mac and iPhone is very convenient, and most of my contacts have iPhones or Macs. So today I almost don't use jabber. Sad but true.

>I haven't found a way to turn off annoying dock animation of Adium

Yeah, my one peeve with Adium. I ended up going into the application package and replacing all the PNGs that make up the animation with the first one. Problem solved.

I love it for iMessages. The ability to read, and reply to all my iMessages from OSX is brilliant.

It's Jabber support is horrendous though. I've been getting messages to it on time, not at all, or delayed. Depending on its mood.

Uncharitable summary: "We did a mockup that merged the designs for Messages and Skype, and we want that interface for Adium. Someone please implement it for us."

I for one totally hate the layout of iMessage, and skype ... which this seems to be a clone of.

I always find myself with both iMessage and Skype having to always check I'm in the correct chat window, this layout to me doesn't automatically say "Hey you're talking to Mike", I have to stop and visually verify each time.

Looking at his design, with almost zero contrast, it would be orders of magnitude worse trying to work out which person you chatting to at a glance.

I would honesty stop using adium if this become the default UI.

Perhaps this is just me (design tends to be a pretty subjective thing), but I fail to see why it looks terrible. Multi-window makes sense for a small chat window on top of other applications. While the chat window itself could use a little work, not everyone (myself included) likes the iMessage word bubbles. However, making this look pretty should simply be a skin for Adium.

the silly chat bubbles is the primary reason i never liked ichat or messages, or the sms handling of iphones.

i was raised on IRC though, so ymmv.

I'd suggest implementing this as its own app—Adium, Pidgin etc. are just front-ends to libpurple, and this could probably be implemented as such, too.

I made a ticket [1] about exactly that with a mockup (which I seem to have deleted from my droplr over that time; though it looked very similar to how iMessages looks now) 2 years ago and the response to the idea was a "no". You should make a ticket yourselves and see if they've changed their minds yet.

I'd love to see a single window UI in Adium but just making the mockup and saying how good of an idea it is doesn't seem to be convincing enough for someone to actually tackle the project.

[1] https://trac.adium.im/ticket/14843

Looks awesome! But not not actually reborn, more of just a concept of how to make it a lot better.

Can we, as a community, stop saying "LOOK I MADE SOMETHING" when it is actually just a landing page or mock up?

I was an Adium user for many years but frankly I very much disagree with Messages being bugridden, while assuming Adium is not. In fact Adium took so much work in trying to update it, logging me out randomly of AOL (and then telling me I am logged in on two devices), not showing things on MSN, not transferring images properly etc.

Much of the crappiness of Adium is a direct consequence of the crappiness of the protocol. MSN is particularly egregious - it is a protocol with almost no structure.

I've written (simple) MSN clients before, and the protocol (or what passes as one) is shocking. It's very clearly a protocol originally intended to exchange strings that has been overloaded with over 15+ years of new features and corporate strategies.

MSN support in Adium comes from libpurple, whose MSN support has always lagged behind the official client due to backwards-compatibility-breaking protocol changes, lack of documentation, and difficulty in reverse engineering. Frankly, I'm surprised MSN in libpurple works as well as it does.

File transfers (particularly MSN transfers) is also a crapshoot due to both bad, undocumented protocol design as well as the base technology. You're running into classic NAT issues that the protocol was never designed to take care of. Who you're sending to and what kind of router they're behind matters a lot.

tl;dr: MSN support in Adium sucks because the MSN protocol is awful.

    >  I very much disagree with Messages being bugridden
• Using search hangs/crashes the client

• Having the client open for more than a few hours will use many gigabytes of memory (I'm at 2.5GB right now)

• Messages aren't in chronological order

• Messages randomly shuffle depending on network conditions (this is the biggest issue)

• Messages randomly don't appear, appear late, or spam notifications on arrival

• "Deleted" messages aren't, so the /Library/Messages/ folder fills with junk over time

• The dock badge will sometimes be wrong, displaying a number when it shouldn't, or even showing negative unread messages

• You can render anybodies Messages.app useless with a bunch of unicode messages. They have to delete their databases and start over.

It's literally the worst application I've used, and you're the first person I've encountered that doesn't feel the same way.

• Having the client open for more than a few hours will use many gigabytes of memory (I'm at 2.5GB right now)

  $ ps auxw |grep -i Message
  jordan   73227   0.0  1.4  4037928 113476   ??  S    13Apr13  13:21.86 Messages*
I've had Messages open for 17 days now and it is using 111 MB of memory (that's RSS, VSIZE is 3.85 GB). I have 4 accounts (three jabber + iMessage).

I don't know much about your other problems. My biggest problem with it is that it doesn't reconnect always when I start/end my VPN sessions, but other than that seems to work ok.

* Edited so it would display nicely.

So I decide to try Adium again after reading this thread, and quit Messages:

  panic(cpu 1 caller 0xffffff800d2a2a6a):
  BSD process name corresponding to current thread: Messages

You do have a point there. No doubt about it. I should have written my comment differently. While yes, I do agree Messages has issues and launched prematurely (and hasn't been updated enough to fix all those issues), Adium has been out for many years and so many problems I've been having with it have never been addressed. So - while yes, Messages is buggy, I'd say no Adium is not a more stable and productive replacement for it (in my humble opinion).

The only issue in the list above that I have experienced is the poor Library/Messages folder usage and that doesn't bother me at all.

I use Messages for everything and no longer use any other IM client - however - if Adium supported iMessage I'd totally go running back to the duck.

I think that might be where the difference is, it's the iMessage service simplimentation itself that is nasty — I've never used Messages.app to connect to AIM or Jabber.

I'd say my no longer using Adium is less about its appearance and more about the fact that it doesn't support iMessage.

Messages (barely) supports iMessage, to the point where I use it to talk to two people and only because they're not reachable otherwise. Big fan of how Messages decides to lose history seemingly at random and only display fragments of previous conversations. Bigger fan of how the conversations it decides to show at the bottom of the log are from weeks prior.

I use Adium for everything else because I know it works.

Agree, so does Android

This seems like adium reborn as skype to me. I've never understood why the current design trends intend to make larger more robust interfaces to simplistic programs like chat. My ideal chat is something like irssi. It takes up minimal screen real estate, which is priority #1 for me.

Looking again, I'm even more flabbergasted that this is getting any play. The UX is terrible.

* It copies the conversation mode in Messages...a terrible feature that ruins the convenience of keeping open only the contacts you want to keep open. Its still multi-window...unless you plan on making me search for contacts. Because typing when your hand is on the mouse is convenient. Nice that I have to click on a contact and bring up their chat window to see their status.

I'm irritated because there is nothing innovative or informative about this post.

Personally I think Adium needs a lot of work, both in UI and functionality. I've used it for a long time, but I'm not particularly happy with it without going out of my way to extend it with add-ons, especially when it comes to IRC. Would love to help on a project like this where I could (design, etc).

There aren't enough designers in the open source community, so I love seeing this effort. If someone implements this, perhaps it will encourage other designers to offer suggested UIs to other open source projects (some of which could really use some UI love).

This would benefit the open source community a whole lot more if it weren't made for a proprietary OS.

Adium has pretty extensive skinning options- I wonder whether you wouldn't even need to rewrite.

It would have to be quite a serious fork, adium as it is has quite a lot of Mac-isms. Starting with it being at core a mutli-window application. I'd be interested to see where this goes, and have set myself a reminder in three months to see if there's been any development on it, as it's an interesting project.

Adium is one of the rare apps where it makes sense for it to be multi-window. You might want certain chats on certain spaces only, while having a separate contact list window that hides at one edge of the screen.

I 100% want different chat windows in different locations on my monitor. Work chat goes on the right, personal chat goes on the left.

I do NOT want to accidentally tell my boss I can't wait to grope him tonight because I clicked on a window and started typing without checking I had the right person selected.

The best thing about Adium is that every conversation can occupy its own little window, taking full advantage of OSX's window system. If you need to chat, the window for that specific conversation is right there to look at. Use exposé and you see all the chats at once.

Never do you have to open the monolithic Adium app, then click to the conversation that exists in a different 'folder'(since they use the same concept of Finder's sidebar).

I immensely dislike the concept of single window apps in OSX just because you can't dismiss the parts of the app you really don't care about. You can't see more than one portion of it at once, and you can't organise it on your very spacious desktop.

Ob-title-gripe: shouldn't it be "Adium Reconceived"?

I expect to see the result in about 9 months.

That's a really nice mock.

I started prototyping something along these lines a couple months before Messages was announced. The idea was to make a (commercial) next-generation IM client for the Mac with a more modern UI. I stopped working on it when the first Messages beta was released, figuring that it added most of the UI things I wanted, and that if it shipped with OS X for free, few would want an alternative.

Now I wonder if dissatisfaction with Messages is high enough that there's an opportunity again?

If you're interested in this stuff and want to talk about it, drop me an email.

That layout will fail if both sides write many short single line posts. You will have to alternate between both sides of the screen all the time and might lose the line on the way.

I never got the appeal of the "left and right" movement. I much prefer messages being shown top to bottom without me having to look left and right. I am not updating my SMS app because they switched to that layout and I found it annoying to use.

I also find a timestamp on each message useful, much more than whitespace or bubbles.

Please make it happen. Also, please let me hide "buddies" that are in my Google contacts, whose activity I do not want to see but whose contact info I do not want to lose.

I think he means "Messages" UI with different colors.

Adium and Transmission are the only two software that I really miss from OSX. I wish someone would make an equivalent on Windows.

Adium, Transmission, Textmate and Transmit are easily my faves.

Main reasons I stick with OS X, despite disliking the direction Apple is taking it.

Mockups are just an abstraction of the real thing. There are lots of issues with this mockup that wouldn't translate well when it gets built.

Example: the search "box" running directly into the pill buttons.

It's a decent mashup of Messages UI with Adium functionality, but that doesn't necessarily make it better.

What is the issue with the search / pill ? It can be done..

I've seen a few of these unsolicited redesigns pass through HN, and design specifics aside I like the tone of this post a lot. Who knows how this is going to pan out, but it's a good place to start.

Yes. Go forth. Fork.

I'm curious, I know Adium uses a webview to implement their chat window. How would one do this without a webview? that reacts the same (text selection only)? does Messages use a webview as well?

I made a HTML/CSS3 version out of it: http://codepen.io/FWeinb/pen/iwIBK

No mention about Skype? I'm stuck on this due to business associates and customers demanding it for the past 5 years.

Everything else is Messages.

I like it:


(nice demo site too)

how was the demo site done? are they using some sort of framework for the showcase?

Looks like he used Adobe Edge.

I don't like the mockup UI he's got on that page. Copied way too much from iMessage which I hate.

I switched from Adium to Messages back when the beta was released and haven't looked back.

Evidently, the UI concept should support both single and multiple window usage patterns.

Hey, I like that, seems pretty... clean. I would totally use that if someone made it :)

Does no one on Mac use Trillian?

The mac trillian is pretty ugly and history browsing is terrible, but I use it. I use it for one single reason: it's the only chat client I've found that can sync between multiple devices. I fire it up on my mac and on my android phone, and not only does it give me the same chats on both, but it's also really good at figuring at which device I'm at and sending notifications there (for example, if I'm working at my desk it does not notify me of new chats on the phone, only on the desktop).

To me this is the killer feature. I would, however, switch in a heartbeat if there was a more featured and attractive app that could do this.

I use it because I loved Trillian on Windows but Trillian on mac is still a buggy, alpha product.

One thing that's incredibly hard to get right with modern IM is intelligent multi-device routing. It's incredibly frustrating to have your phone buzzing next to you as you're having a conversation on your PC. Using Trillian on all my devices prevents that and that's reason enough since Adium doesn't have a iPhone app.

I've been using a Mac since OS9 and I didn't even know there was a Mac version of Trillian. I thought it was a Windows-only client.

what's buggy about iMessage? I haven't had too much trouble with it, at least for SMS and gchat. Then again, I usually use Adium on account of IRC so maybe I just haven't used iMessage enough to run into anything.

I think he means iChat

Perfect. Love it.

looks 10 million times better, nice work

Please do.

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