The Future of Transmit iOS | Hacker News Reader

The Future of Transmit iOS

panic.com

138 pointsDoubleguitars126 comments

I hate to admit it, but I have both Transmit and Coda for iOS and I never use them. I've never really used them. I know a lot of care went into making these applications, so I don't want to say it was a waste of money, but I just can't make iOS any more of a productive platform... even with great software.

Honestly, this story/link should be considered maybe a both a blow to the "post-pc" narrative (we're going on 8 years of "post-pc") AND to Apple itself.

MS didn't go about Win8 very gracefully, but honestly, I think A LOT of the thinking behind Win8 was right-on and future-minded. These concepts just needed a gentler, multi-staged introduction. I think Win10 is turning things around.

Also: I wonder if this reality has anything to do with that "cross-platform" UIKit/AppKit replacement PR drop we got a few weeks ago?

Which iOS platform, iPhone or iPad?

The newer iPad Pro model with the keyboard cover seems like it would be much better for productivity tasks than an iPhone, for example. Still nowhere near as functional as a laptop for day-to-day productivity - but maybe good enough that you could bring an iPad along on a trip and make a quick edit or restart a server if needed, and leave the laptop at home?

Both. I had a first gen iPad Pro for about a year before I gave it to my parents to use for their travels.

The keyboard cover was super handy (I highly recommend it), and I admit that there are some cases where I could have gotten away with just an iPad. But for anything important, I'd need to open a VPN and RDP session anyway. A Surface may have been a better purchase, TBH.

VPN and RDP is a task you can do easily on iOS (VPN is embedded in system and Microsoft RDP client is very good, and free). There are serious real limits in using iPad+keyboard for work, but not VPN and RDP.

Agreed on that! The iOS RDP app is pretty darn good, and VPN connectivity options are really good too. BUT... I discovered that if I RDP into my work machine/network, it totally negates the need for both Coda and Transmit on my local iPad.

Thats the trick... :(

My son decided to hand his laptop to his brother a few months back and try and run his entire music business using just an iPad Pro. It works, up to a point, but there are just so many niggly issues compared to just using a laptop.

Simply things like a booking agency sending him a PDF contract to sign becomes a saga of jumping between multiple apps, importing and exporting etc. in order to get it done and sent back. Sending sound clips to his customers is also surprisingly difficult, as there is no easy integration between services like SoundCloud and DropBox, and even with the new 'Files' app, it is difficult to manage more than a few files in a complicated folder structure and send them on to others easily.

He also found it a challenge to make simple posters etc. for his upcoming shows and send them to printers or even to Facebook. Supposedly simple acts like adding readable text to a photo background was a challenge, and he ended up going through about 6 different apps before he found one that worked the way he wanted to.

Add to that the fact that his iPad Pro pen started to fail after a couple of months, and I think he has regrets going down this path.

Careful. People on HN aren't really a part of the "post-PC narrative". People plugging keyboards into iPads and using them full-time to write papers at school are.

I'm not really clear why I'd use Transmit on the Mac either, except that it's a very pretty interface.

> People plugging keyboards into iPads and using them full-time to write papers at school are.

How many people are doing that? A chromebook I can buy as a school computer. An iPad... The idea of trying to select chunks of text and move them around in a paper, or simply re-positioning the cursor sounds like a nightmare.

If I were a kid I'd beg my parents for a real laptop.

That’s because your idea of a “computer” has a command line, Python interpreter, filesystem, etc.

Kids these days don’t get exposed to that so they don’t really desire it. Their idea of computing comes from apps that are take-it-or-leave-it, interact with other apps only in predefined and constrained ways, and use touch UI. They may want a better text selection interface but I’m not sure they concretely know that a “real laptop” has one.

But isn’t that a failure of their education system?

Compare the kids who grew up on their iPads against a kid in Eastern Europe who grew up hacking on a Linux or even Windows machine using a real keyboard. Wouldn’t the latter have all sorts of advantages in all sorts of jobs?

If nothing else, growing up using a real keyboard would mean they could type a lot faster than the one who is used to a touch screen.

>Compare the kids who grew up on their iPads against a kid in Eastern Europe who grew up hacking on a Linux or even Windows machine using a real keyboard. Wouldn’t the latter have all sorts of advantages in all sorts of jobs?

No, especially since the other kids grew up in a more dominant economy and have way more opportunities (which matters more than learning to mess on a cli).

Besides, the degree to which Eastern European kids learn to "hack on Linux" is perhaps overrated. Most kinds in school couldn't even learn to power on the PC on their own (but amazingly one would teach another all there is to start playing some game).

No, because it takes precisely 18 minutes to use a keyboard if you already know how a touch tablet works, and 3 days to learn to touch type.

So, they'd have a 3-day-18-minute advantage in all sorts of jobs

My idea of working on a computer involves the ability to at bare minimum select, edit, and copy and paste text without struggle. As does any kid exposed to one of these 'magic' devices, of which I can assure you they ARE exposed.

Attaching a keyboard to an iPad does not make those tasks significantly easier in iOS, unlike a Chromebook - a device which outsells the iPad and continues to grow each year.

K-12 Sales Chart -> https://d3e7x39d4i7wbe.cloudfront.net/uploads/photo/image/53...

Keyboards have arrow keys, and iOS supports shift-select and the Mac clipboard command keys, so you might give it a try if you’re having trouble with the touch interface. Kids probably don’t have the same issues.

> Kids probably don’t have the same issues.

Your argument consists of imagining that kids prefer iPads to complete certain kinds of work when the data tells a different story.

Chromebooks are vastly outselling iPads and their margins continue to increase. iOS has functional shortcomings at productivity tasks, and those functional shortcomings don't evaporate if you happen to be a kid, unless you just want to repeat the party line painted by Apple's advertising.

I'm not a believer in the magic superiority of kids either, but neither do I believe that sales numbers are a reflection of specific productivity features. Who makes those buying decisions based on what?

Anecdotal, but my kids have school-issued iPads and personal chromebooks. They prefer the iPads for pretty much everything (except perhaps typing long papers). Watching them edit a document on the iPad drives me nuts, because the select-copy-paste process is so inefficient... but it's how they prefer to work.

https://itunes.apple.com/us/app/pythonista-3/id1085978097?mt...

A little pricey perhaps, but great.

Given how good Pythonista is, and how much work the developer puts on it (I’m in the beta program and there are very frequent releases, and he’s very active on the forums for the app) is well worth it if you use or like Python (I don’t even write that much Python anymore, and less on iPad, but when I do it’s on Pythonista). Aside from being on the beta (which would mean I’d have a free copy of the app if I hadn’t purchased it) and some tweet exchanges, I have no relationship with the developer.

>Kids these days don’t get exposed to that so they don’t really desire it.

These days? Kids don't get exposed to those things ever since Windows 3.1, with the exception of the filesystem.

Honestly? For writing, for me at least, there's literally nothing better than iPad Mini + Magic Keyboard (no plugging needed) + smart cover.

It's nearly a third the weight of a MacBook Pro, you can stick them in your jacket pocket (the iPad Mini is barely wider than the keyboard), the keyboard is full-size and you can rest it on your lap for ergonomics, and as long as you're just dealing with writing text, the screen is plenty large enough.

(And the keyboard lets you do all your cursor positioning with the arrow keys, literally no different from a laptop.)

For programming forget it... but for writing, it's the best thing I've ever found.

I find that hard to believe. What about basic text editing like selecting and moving around blocks of text, copying and pasting between documents, etc?

Those kinds of tasks are excruciating to perform on iOS.

They’re really not at all painful on an iPad with a keyboard. I’m going to guess that your experience is based on a much older version of iOS, but it’s been quite nice for the past few years (and it’s evolving still). Another option: perhaps you’ve only ever learned to do those things with a multi-button mouse? For my experience, I’m extremely comfortable with a variety of keyboard-based systems (especially Emacs), and the iPad interface is totally usable.

> I’m going to guess that your experience is based on a much older version of iOS, but it’s been quite nice for the past few years

I have an iPad and iPhone running iOS 11. I'm not alone in thinking that what should be a basic feature like text selection is awful.

From 10 days ago in the iOS subreddit, "Text Selection is Broken in iOS": https://www.reddit.com/r/ios/comments/7mcvhu/text_selection_...

> How can something so important & basic literally be broken in the most recent version of iOS?!

> It’s been like this for what seems like forever. Having to select some text an upwards of 10 times is pretty annoying.

It seems other people feel the same way.

Reddit thread you linked is about basic text selection via touch.

GP is talking about text selection using an attached/bluetooth keyboard.

It shouldn't need to evolve, it should just work well and it always should have.

Why would those tasks be harder to perform on iOS than on a Mac if you have a keyboard on both ?

Mouse

I would be amazed if selecting text on a mac with a mouse was quicker or easier than selecting text on an ipad with a keyboard.

I… wouldn't. I'm so much faster doing text operations on a Mac, it's not even close. A pointing device is just so much more quick/precise for me than fingers.

That's what I don't get about these arguments in support of text editing on iOS. The mouse isn't bound to the physical dimensions of the screen. It also doesn't require you to put it atop/obscure the screen to work, either (unless you're doing something very wrong).

I'm sure there are zippy ways to edit text on iOS, but like you, after using a mouse, hell, even a trackpad, I think it's soooo much quicker using the latter.

1 quintillion upvotes

Text selection on iOS can indeed be a pain, particularly in Safari where it seems to lock onto html divs ignoring your selection and jump to the entire paragraph.

On iPads and 3D Touch iPhones things are generally a lot better outside safari and when editing text. You can 3D Touch or two finger poke the on screen keyboard to activate a text cursor and even select text by pushing again with 3D Touch. you can do the same with an iPad but it’s a little more clunky requiring an extra tap.

Of course you can’t beat a good old text editor.

In essence the iPad is made for people who never learned how to use normal computers.

If you are a person who would mash the delete key once per character to delete a line of text (e.g. my parents) then perhaps the excruciating text selection is not really a negative, as text editing on computers was "always" excruciating. The simplified user experience, is however a positive.

If one really took the time to "learn to use normal computers", then they would be perfectly capable to use a table+keyboard-only interface (and keyboard shortcuts) to do all kinds of editing tasks without even taking their hands from the home row. No need to "mash the delete key once per character to delete a line of text", at all.

So your argument is more true for middling users than for expert users of something like vim or emacs that can do every text edit function quickly with an external keyboard in iOS.

> expert users of something like vim or emacs that can do every text edit function quickly with an external keyboard in iOS

Show me where iOS implements incremental search and a persistent mark, then. For that matter, show me where any OS implements them.

You can run vim itself on iOS. As well as any number of other editors, and tons of standard and custom shortcuts are available -- to the point that not having access to a mouse for text selection is not any big deal for writing.

My experience with a keyboard on iOS has been awesome. Even more impressive is that large handful of the ⌘ keyboard shortcuts work, including some of the globals like ⌘-tab (change app) and ⌘-space (spotlight).

Selecting text on an iPad is a breeze if you use what Apple calls trackpad mode. Resting two fingers on the Keyboard lets you slide around the cursor. A double tap selects the current word. Resting your fingers before moving initiates a selection mode instead of moving the cursor.

Multi-document work is a pain though. Simple school tasks such as reading a paper, taking notes and creating a short essay or presentation from your notes is not a great experience on an iPad.

When I went to the boy's orientation last summer at UIUC, a lot of the parents were doing that --- like, maybe one in every other row. That was the moment it sunk in for me that Apple might have been right about this (if not right about the timing).

"How many people are doing that?"

Probably very few. AFAIK, wired keyboards were never officially supported. Plus one needed the "Camera Connection Kit" adapter (later rebranded as Lighting to USB connector). Plus some USB keyboards would not work.

The only officially supported keyboard option was bluetooth keyboards.

In the real world, I have never seen anyone plug a keyboard into an iPad. I have only seen people use bluetoth keyboards.

Apart from looking nice it works much better and more intuitively than, say, Cyberduck

I’ve found coda to be vaguely useful (and prompt to be very useful) but can’t imagine what I’d even use transmit for.

>Honestly, this story/link should be considered maybe a both a blow to the "post-pc" narrative (we're going on 8 years of "post-pc") AND to Apple itself.

Post PC can mean two things, neither of which is invalidated by the story:

1) Tablets, smartphones and so on, being increasingly used by regular people (in lieu of PCs) to do work (writing documents, using work apps on the move -- e.g. airport personnel, factory staff, graphic designers doing digital painting on tablets, etc).

This hasn't particularly caught on, but it doesn't really concern the developer crowd and its more specialist needs. Everyone could very well use a mobile phone to write a work paper today, but nobody expected or predicted phones to have a full terminal/file transfer/scripting/programming etc workflow to satisfy the average developer/admin. That's not a real part people though as the "post-PC world". This is usually conveyed in the "Car vs Truck" dichotomy (I think invented by Jobs), where in the Post-PC world mobile phones are like cars (for regular use) and PCs like trucks (specialized use cases).

2) The other sense of which people speak of a post-pc world is tablets, smartphones and so on, being increasingly used over PC (in total hours/day) in general. Which is true and getting truer everyday -- be it for web, mail, social media access, video, etc.

Transmit for iOS is $10, but I bought it because Panic is known for making premium-quality software to match the price.

…So I was quite surprised to find out it doesn't support the SMB protocol (i.e. Windows file sharing). That's one of the most common file transfer protocols, not to mention the only one that current macOS can serve out of the box. And the lack of support is not an OS limitation, since there are other iOS apps that support it (e.g. Documents by Readdle). So what gives?

I suppose the lack of support is inherited from Transmit for Mac; macOS can mount SMB network drives natively, so there's no need for a separate app. (Although I might like to use one anyway - mounting network disks is always a laggy affair, since most software doesn't expect the filesystem to be that slow.) But iOS has no builtin SMB client, so it would be quite useful there - is useful, with those other apps.

The list of cloud services supported by Transmit for iOS is also rather small - S3 and DreamObjects, that's it - even (especially?) compared to Transmit for Mac, which supports something like 10 other services![1] Why couldn't they port that functionality over?

The app does have a quite nice UI, and unlike some, I place a high value on that. But a premium UI is no use if I can't even connect to the servers I'm trying to access. Maybe they'd have gotten more sales with an app that could.

[1] https://panic.com/transmit/

Why would an FTP client support SMB?

I mean yes, its obvious that this is probably a good idea for a phone - I think its one of those things that wouldn't occur to anyone to do.

Well, if you think of it as an FTP client. But it’s a multi-protocol file transfer client; I mentioned that the list of protocols was short on the iOS version, but it still includes three options that don’t have “FTP” in their name.

I know what you’re saying wrt “wouldn’t occur to anyone to do” - true, desktop file transfer apps don’t traditionally include dedicated SMB support, since the OS can handle it. But the idea did occur to the developers of multiple competing iOS apps, including the one I mentioned, which support SMB in addition to FTP, WebDAV, and various cloud services.

(FWIW, I’m not claiming I was scammed or anything like that. The feature list was right there in the App Store description, and even after buying the app I could have tried to get a refund. I didn’t, because I have use cases for it other than SMB. I’m just surprised that it’s missing.)

I think of the UI one would use on an afp or smb connection as very very different than the one used for ftp - I suspect thats a me thing, rather than a reality.

I'm not really sure why you would expect Transmit for iOS to have SMB, and in their product website and app description, they clearly outline what protocols they support. FTP and SMB are completely separate protocols?

See my response to someone else with a similar question:

https://news.ycombinator.com/item?id=16083772

> I suppose the lack of support is inherited from Transmit for Mac

I'd add that CIFS/SMB is one beast of a protocol... a basic ftp client for up/download can be done in a day, but CIFS/SMB? The spec alone clocks in at healthy 441 pages. Either reimplement a huge subset of this (and test with the countless versions of Windows + the even more countless deployed versions of Samba) at considerable expense, or ship Samba with the app... there are IIRC some Android apps doing exactly this, but I don't know if it's at all possible to do so on iOS.

Samba is GPL, but there are other options. I Googled 'smb client library' and saw at three viable-looking libraries on the first page:

https://github.com/naxos/SMBClient

> SMBClient is a small dynamic library that allows iOS apps to access SMB/CIFS file servers.

https://github.com/videolabs/libdsm

> lib Defective SMb (libDSM) is a SMB protocol client implementation in pure old C, with a lot less features than Samba but with a much simpler, and a more permissive license

[…]

> The initial goal of this project is to have a library that can access most SMB shares to read files and that has a license compatible with the iOS/Android/WinRT appstores in order to integrate it into VLC for iOS and VLC for Android.

https://jcifs.samba.org

> JCIFS is an Open Source client library that implements the CIFS/SMB networking protocol in 100% Java. […] This client is used extensively in production on large Intranets.

(Java would be annoying to run on iOS, but probably doable, perhaps using Google's Java-to-Objective-C translator.)

There's also Apple's own SMB implementation, which used to be included in opensource.apple.com source drops, although it seems to be gone as of macOS 10.11; here's the latest version:

https://opensource.apple.com/source/smb/smb-759.40.1/

Or you could go to the BSD SMB implementation that it was based on. I'm not sure how difficult it would be to port either of these to run as a standalone library - it looks like parts are already a library, but smbfs itself is a kernel module - but it's probably viable. In fact, 'fs-utils' from the rumpkernel project, which is an existing effort to port filesystem modules from the BSD kernel to userspace, seems to have a version of it:

https://github.com/rumpkernel/fs-utils/blob/master/lib/exter...

…Forgive me for the wall of links. I guess one example library would have been enough, but I was curious what options were out there.

I don't know how these libraries compare to each other and to Samba in terms of quality and compatibility. Most likely Samba is the best, but they all seem widely used enough that I'd expect them to do the job in most cases.

Every time I read stories like this, I feel like the value of iOS development is slowly dwindling. App development has been a race to the bottom for small teams/indie developers making it difficult to build a sustainable lifestyle business. I'm not sure if there is anything that can turn the ecosystem around, but I hope WWDC introduces some more sustainable plug and play business models for developers.

I never understood the needs for the vast majority of apps. With everything clearly moving towards the web over the last twenty years, why did we all of a sudden focus on native applications? Money is the only one I can think of. I sure as shit don't need a native application to view news / tweets / whatever other nonsense there is these days. I've limited myself to the very first "page" of my iPhone. No more.

The web of software has inherent unsolveable security issues and leads to an ecosystem where less and less software lies under the direct or indirect control of the user.

Proprietary apps are not the solution to this either, of course.

We need native, free (not necessarily without exchange of money) and open source programs that do the work for us.

> inherent unsolveable security issues

And UX issues, and performance issues.

Building (and using) native apps over "web apps" for non-trivial tasks is definitely the right way to go. And, no, I'm not talking about apps using Electron or the like.

Because the kludge of HTML 5, CSS 3 and JavaScript is no match for actuall good UI/UX and platform integration, as proven by all WebApps platforms introduced since the WebRuntime on Symbian OS.

WebAssembly will be the final nail, making the browser just yet another general purpose VM, targeted by all major language eco-systems, using their UI framworks on top of Canvas and WebGL.

Admittedly, I have barely used MS Word in at least a decade, but I find Google Docs a far superior overall experience. Do you honestly have no positive web app experience at all?

Google Docs is at the level of Word 2.0 for Windows for Workgroups. Aka an improved typewriter version with some support for images.

No, I have yet to use any web app that could provide a better UI/UX than native alternatives, specially in execution speed.

So you'd rather Hacker News was an app correct? I sure don't.

I use the Reddit app yes.

Better yet would be for HN and any other kind of discussions forum, to just be NNTP hosted and then I could use any newsreader I like.

The Internet is about network protocols and distributed computing, not trying to shoehorn everything into http://

www would disagree.

Mobile native apps and the market failure of all browser based OSes would disagree.

Even ChromeOS only had an impact in the US school system, forcing Google to support Android native apps.

Just because browser based OSes haven't been spectacularly successful, doesn't mean that webapps as a whole are failures. I am willing to bet that Google docs has better usage on the desktop webapp than it does on the mobile native app.

I agree with you, the vast majority of apps aren't really useful or would be just as good being websites. I've culled my iOS apps down to 1 page an a half. The home screen contains apps I use and the second page contains apps I kind of need or can't delete. Everything else got deleted.

The web is poised to make a big comeback on mobile. New browser features and PWA techniques have closed the gap considerably. There will always be a market for real native apps but a lot of people are going to realize that a PWA is good enough and easier to build and maintain.

Not with Spectre exploits on the horizon.

I can almost imagine a future Craig Federighi post to Apple.com, "Thoughts on Javascript"

This is something I'd expect Apple & Google to fix in their JS engines, so I don't see why it's more of a risk for web apps than native apps. If anything it's easier to fix.

Using my phone on a big European city right now, most of the time I get only Edge network access, with luck.

Just browsing HN is an exercise in patience, while I have plenty of native apps to keep me busy.

No, Web won't do any comeback on mobile.

This is the entire point of service workers, which will soon be supported in all major browsers.

A pile of code, needed to proxy network access, that even requires scaffolding utilities to simulate what native code is able to get for free.

The web gives you a lot of things for free that are hard on native. Engineering is always about trade offs and you’re going to see more people opting to accept the tradoffs of the web over native on mobile, just as they have for years already on the desktop.

Not when their UI/UX mobile experience keeps sucking.

Regarding the desktop, ChromeOS is a failure outside US school system, I am yet to see at least one person using them in an European coffee shop or retail, hence why Google has added support for native Android apps.

Yeah I totally agree. I think that if you couple PWA with all of the new Smart Contract Web 3 stuff that's coming, it looks like there is going to be a resurgence.

A large number of apps never really made sense. Transmit for iOS is, in hindsight, one of those apps. When I look at the apps on my phone, they are all apps that somehow tie into a platform. Platforms can be something as big as Youtube or Facebook, but it can also just be your local bank or postoffice.

Indie developers aren't building platforms, they just build apps, that's why they fail. Small teams still play a big role in app development, but as in-house developers or consultants, building apps for bigger platforms.

I can't name one standalone app that's successful, perhaps the calculator? Games are different, and struggling with their own issues, in respect to profit.

Personally I don't see apps disappearing, in favour of webapps anytime soon. Email should have been the first thing to go from apps to the web, and yet everyone I know just use the built in email client on their phone.

I don’t know, I think the iPad as a serious content creation device, something an FTP client would help with, never really materialized. Apple keeps trying but it’s not there.

> Transmit iOS made about $35k in revenue in the last year

Not good given the obvious development effort that went into it, but the transparency is appreciated.

I suspect this is directly related to the extraordinarily difficult experience that comes with trying to accomplish technical work on iOS. Perhaps that will change as iOS matures. Personally I've tried to use even boring old ssh and vim from an iPad a handful of times and it's beyond frustrating, you'll never want a real keyboard and cursor more in your life.

I'm afraid iOS never "matures", except maybe by important user-facing parts of it being consumed by MacOS when it becomes viable on i-devices.

iOS was designed with entirely different use cases in mind, entirely different workflows, and completely different users, those who prefer to never know words like "vim" or "ssh", and honestly have no need in either.

It's a square peg in a round hole.

iOS has matured some and continues to do so. The Files app and drag and drop enhancements in iOS 11 are good examples of this. I don’t know if iOS will ever meet all of the use cases of a desktop OS, but it has definitely become capable of doing a lot more than it could a few years ago.

The thing that actually surprises me most here is that Transmit for Mac is apparently quite profitable. Free FTP clients have been around for so long, and have worked so reliably, that the thought of paying for one never crossed my mind. I personally have used Cyberduck for years, which is free (although it does tastefully ask for donations) and has never given me a reason to look for other options.

Can anyone explain what Transmit does that makes it worth $45? I am wondering if I am missing out on something here.

Transmit/Mac when it's discounted is definitely worth it IMO. $45 is a little steep, but Transmit performs really well, especially when you have a lot of little files to move.

$45/yr is probably an hours worth of work in most of the US for software devs. If you work minimum wage that’s probably 5 hours of work.

It literally needs to save you about an hour’s worth of time over its lifetime to be worth it. And that isn’t even including all the other benefits you get from using a more frictionless and pleasing to use software.

I find complaints about the steep cost of tools from professionals when they don’t even cost triple digits quite short sighted, to be honest (an exception could be made for professionals from countries where their pay is significantly lower).

I'm fine with paying for good software and I probably have about a half dozen paid apps running right now on my desktop. I completely agree with your point that $45 is probably not worth worrying about (or the difference between $45 and a $25 sale price) in the grand scheme of things if the software is useful to you.

One thing I will say though is that there are other costs associated with non-free software beyond just the license cost. You've got to manage license keys. You've got to worry about upgrades. If your company is paying for it you've either got to go through IT for procurement or you've got to submit an expense report to be reimbursed. Sometimes there are obnoxious license servers that you need to deal with (I'm looking at you, IntelliJ!). Installation of paid apps is harder to automate than a simple homebrew or npm command.

Of course, all of this may still very well be worth it if the software is useful.

It’s even more puzzling given how many other FTP/SFTP graphical clients there are for Mac OS. Not just Transmit and Cyberduck, but also Fugu, which is open-source like Cyberduck (although it has not been much developed of late) but Fetch, which is commercial and older than Transmit (dating back from the 80s) yet cheaper at $29.

I just don’t see why Transmit is worth it with all those other options.

Its well designed, supports a good number of protocalls, and the "mount as disk" feature can be really handy both for daily use as well as automation.

I agree 49 is a lot though. I'd wait for a bundle or discount.

We spent the money because Panic is good and a lot of non-technical people are told by the government, as part of grant and reporting requirements, to upload/download files. $45 is pretty cheap to do a setup and have people do it on their own with a very user friendly program.

I paid $34 for Transmit 4 in 2013 and $35 launch price for Transmit 5 in 2017. Before that I used Cyberduck (and donated in 2010 to remove the banner!) I have also used Filezilla a lot on multiple OSes.

Transmit is better software. I mostly prefer it because it is better at managing favorites, I prefer its file editing workflow, I prefer its control over transfers and aesthetically I like the UI and how it presents multiple panes. Its directory mounting story is pretty good. I trust it more when I have multiple transfers in progress, etc. The use of tabs in the 5 update and easier search filtering of favorites has been worth the update so far.

Ideally everything I do would be a git deploy or an rsync command, but since it's not, I am happy to pay a quality software publisher to remove a lot of headache from annoying FTP and S3 operations. It's no different than why I pay for a good text editor or image editor even though I can get anything done in a free alternative.

I used an older version of Transmit years ago (it was a bit cheaper then) and was always happy with it.

Then I went years without needing to use FTP at all (capistrano deployments for the win!)

Over the last year and 1/2 I have been doing lots of integrations with 3rd parties where some varient of FTP is needed (sFTP, etc) to download data.

I tried a few different free ones and used CyberDuck for a few months, I started running into strange bugs that were not reproducible and basically gave up on it.

Downloaded Transmit and never looked back - it just works, also a great S3 GUI client.

For me, Cyberduck was unreliable. It would constantly disconnect from my servers (VPS, shared hosting, Windows file server at work, even my local network servers), it would crash occasionally, and very rarely it would lock up for a few seconds then run fine (relatively) for a while. This was the case both on my Mac Pro running Sierra and my Mac mini running El Cap (I have neither machine now; sold the Pro and the mini's logic board died).

Command line ftp worked fine, Filezilla worked fine, and Transmit worked wonderfully on both machines. Anecdotal I know, but Cyberduck is not what it used to be, at least for me.

Sad to hear this. Transmit for iOS is a fantastic app and a staple for my usage of iOS. Since Apple still hasn't implemented any sane ways to easily share files between iOS and PCs, Transmit is always there to allow a quick safe upload to a trusted server (with no "cloud service" nonsense involved!).

Hope it will keep running for a long time even without updates!

I love Panic, and their software is great quality for what it does, but I'm sad to say I just don't have a place in my life for those kids of development tools anymore.

I find it unsurprising that there isn't a market for Transmit on iOS, but I do find it surprising that there's still a large market for Transmit for Mac. I bought it in 2011, and got a year or two of use, but now only use it once or twice a year which doesn't justify an upgrade.

I'd love to see them focus on more modern development tools, but I'm not sure their approach will work in that context.

might be why they started making games. Firewatch (the tree icon on their home page). It was fun and different, but failed to make me more productive...

As I understand it, Firewatch was not made by Panic, but the founders of Panic had creative input into it. I don't think it's a company focus for them.

Removing, from what I can see, a pretty good piece of software from sale doesn't seem the correct option.

Then again I understand making it free / heavily discounted may irritate people who have bought it already.

Is there a better option for apps such as this?

I can see why they wouldn't make it free, they would have all the support costs without any revenue.

In an ideal world someone would be willing to take it over but who would want to take on a product that is a commercial failure?

It's only a "failure" if the goal is to make money with it. If the goal is to build good will and help their users who have bought their other products, and help recruit new users for those other products, it doesn't have to be looked at as a failure.

They could always offer it without support. Or open source it.

I bet they won't. I think there is a (or maybe a few) soecial and proprietary Panic libraries that they include across their applications. Open sourcing Transmit iOS would also mean open sourcing part of their other conmercial applications.

I did say commercial failure. Good will doesn't put food on the table.

As a developer, I can't eat good will, and my landlord doesn't accept it for my rent payment.

>Finally, the new Files app in iOS 10 overlaps a lot of file-management functionality Transmit provides

I disagree completely. Every time I need to save a file to my phone, I can’t use the files app because you can’t just save files the root, and I can’t make a directory in the root either. You need to save into an App or an App’s directories. So I need a “container”, unless I’ve completely missed the point of the Files app.

EDIT: I should say, this is for the “On My iPhone” option, not iCloud Drive, which I don’t use.

I always save files to the Transmit “folder”. In fact, before files it WAS the files app on my phone, and it continues to be today.

I’m very sad to hear it’s being discontinued. I wish they’d open source it so it could live on in some way. I’d be willing to pay money to help this app live on (e.g. pay the Apple developer fees to host it myself).

Damnit. Are there any good replacements for Transmit?

I love Panic's tools, but I was one of those that got burnt when they pulled their StatusBoard app off the App Store a couple of years back. I have a small side project [0] that I created a specific integration with StatusBoard because it suited our web service so neatly. Shortly after I launched my app, they shut it down!

Haven't been able to find another app like it, and I believe a lot of our customers (including myself) still use an older version of StatusBoard still, so I've left the integration info on our web site for now.

But like others on here, I have most of their other tools on my iMac and my iPad, and while I use the iMac ones semi regularly, I hardly ever use the iOS versions (apart from StatusBoard).

[0] - http://www.staffstatus.io

As someone who really wants to use my iPad for everything, I really really tried to use coda for development. But it crashes soooo much. It’s sad because the potential is there. The iPad is so powerful but we need apps to utilize it fully.

Also, relatively unrelated, I await Apple to release Bluetooth mouse support. The touch screen ergonomics aren’t good for long perdiods of hackin:-)

Aw that stinks. Transmit was the first app I bought for my iPad Pro. I use it every day.

I guess it'll keep on working for now.. hopefully someone else fills the gap in the future.

I simply had no idea it existed. Bought just to support a great developer, but also because I may find uses for it in the future.

I still can't get into tablets but I can absolutely see how this would make tablet work feel more like "real" computer work.

Same here! I honestly would have loved this a couple years ago, but this is the first I am hearing! I have been using Transmit on Mac for well over ten years.

Try FileExplorer. Not as pretty, but substantially similar at 50% of the price.

Bummer. Coincidentally, I just (about an hour ago) purchased a bunch of license for Transmit 5 Mac -- fantastic product. I've also been a Prompt and Transmit iOS user since release -- also great products.

That's a real shame. I'll keep using it, but at some point I guess I'll need to find a new SFTP client. Any recommendations? Bonus points if it integrates with the new Files app.

Also sad to see it go. I use it all the time to transfer videos from my Linux workstation.

Sad to see it go!

Crappity crap... Then again, it was very rarely updated - usual sign of noncommittal developers.

> Crappity crap... Then again, it was very rarely updated - usual sign of noncommittal developers.

I call bullshit.

Panic has been developing apps for the Mac for many years. Wikipedia lists Transmit for Mac as first released in 1998. Transmit 5 was released several months ago. Though they have been slowly ending development on their iOS apps, this is a preferable outcome to burning money and closing up shop altogether.

Disclosure: I am a customer.

I was talking about Transmit for iOS specifically.

> usual sign of noncommittal developers.

Or stable software. Not every application needs monthly updates.

It’s fairly different for iOS, where base OS requires at least yearly updates to keep up with the times (and at least with iOS 10 such update was very late).

This does not conflict with anything I said. Yearly updates != monthly updates.

And even then you don't always have to update for each new iOS release unless you intend to make use of new features.

I had an app on the AppStore for 3 years without updates before Apple said 'this app needs updating or we will remove it from the Appstore.

Wow, I just saw the price. $10!

There seems to be a lot of entitlement amongst developers of successful apps. I mean ten dollars... Have they ever considered that it could be a free app that is a thank you to the community that has supported them over the years, which could pay for itself by bringing new users into their customer base?

I think the entitlement is on your side, not Panic's.

Panic are a business and to provide and maintain the app, they have to assign one developer half-time. This costs more than the $35k or so earnings of the app.

I'm not sure the math could be any clearer.

They earn plenty of money on Transmit for Mac.

The following may be a bit hard to argue but let me try. The availability of Transmit for iOS should increase sales for Mac, since the apps can have synergies that make the Mac app better. So I think they should credit some of their Mac earnings to iOS, and factor that into their figure for the iOS app. Doing separate accounting for paired apps on different platforms makes less sense when those apps have synergy with each other.

>So I think they should credit some of their Mac earnings to iOS, and factor that into their figure for the iOS app. Doing separate accounting for paired apps on different platforms makes less sense when those apps have synergy with each other.

I think they disagree with you here, bud.

Yes, either that, or it may not have even occurred to them to think of it that way.

> Or a free app that is just done as a thank you to the community that has supported them over the years? What would be wrong with this?

That would be a very expensive thank you for a tiny business! Given Panic's level of polish and concomitant development timelines, I expect this app took several months of a designer's time and at least six months of an engineer's time—probably more like nine. Then QA, support, etc. I'd guess we're looking at a >$150k price tag here, plus ongoing maintenance.

As to paying for itself via bringing new users into their customer base: given the average sale price of their desktop software, this iOS app would have to generate several thousand net-new sales. That doesn't seem wildly outlandish to me, but I'd guess one order of magnitude fewer is more likely.

>That would be a very expensive thank you for a tiny business!

I wonder why they don't just scale back the maintenance to bare bones and make it free. The past cost is already behind them... removing it from the store won't fix that. They even did the iPhone X compatibility updates already. For the new protocols and cloud services they'd like to add, I could see an intern tackling that, for a relatively lower cost (not free) compared to a developer. So I'm not so sure the numbers couldn't work out.

You haven't responded to the point that the cost is sunk.

I have made free apps and the indirect dividends have been huge. You yourself make a free app (and I donate to it monthly, hurray for me) and yet you don't see why anyone would want to make a free app?

Cool. I think you're entitled for charging your rates for your labor. Do free work for me.

I've done plenty of free things in my career and in almost every case it's turned out to be a huge win, because of the relationships it led to, doors it opened, etc. Not everything good in life makes money directly.

Reply on news.ycombinator.com