“Git Tower” revokes licenses when discussing bugs/feedback on Reddit | Hacker News Reader

“Git Tower” revokes licenses when discussing bugs/feedback on Reddit

reddit.com

68 pointsValentineC41 comments

> Hi, Julian from Tower here! We were just made aware of this post via email. We would never revoke a license unless a user requests a refund. We highly value each form of feedback to our app, especially critical feedback. Thanks to that feedback, we have been able to improve Tower over the years and continue to do so!

> It would be great if OP could reach out to us, so we can find the corresponding license and check why it seems to be no longer working. Thank you!

https://www.reddit.com/r/git/comments/ao1q7t/care_git_tower_...

Yay, another case where people assume the worst possible interpretation of a situation, posts about it on the internet like it's fact, and an angry mob forms. What a wondrous time we're living in! /s

Seriously, I view this as horrible behavior on the reddit OP's part. Whether it turns out to be Git Tower blocking them or some other problem, not taking the time to try to figure out what happened before publicly asserting something is inexcusable.

So just to clarify: It is horrible behaviour on the OP's part to make assumptions about other people's motives, but you're going to go ahead and make assumptions about the OP's motives just the same?

Seems a little hypocritical.

I'm making a statement about what the OP posted and provided, which did not include any indication of contacting the company in question to find out what hapoened, did not include any info on how that company would have known what license to revoke, and in a reply to a company spokesman on reddit also did not note any of this info.

If the OP did have any of this info, not including it is also bad. My problem is with the OPs observable actions.

To be fair, this could also just be a PR response to diffuse the situation. You don't really know who is being honest here.

I'm only responding to how one side is asserting intention on the other side with the only evidence being an outcome, and no note that any investigation was done.

It would be irresponsible for me to post that my neighbor defaced my house because I badmouthed him in the neighborhood immediately after I notice the defacement but have no evidence other than my assumptions. Whether the neighbor is guilty or not does not absolve me of my responsibility to back up my assertions if they are made as statements of fact, nor to look into the situation and provide what evidence I can before heavily insinuating a beleif.

These are of course my own moral and ethical beliefs, but since I try to practice them myself and I think they make the world better overall, I feel comfortable in expressing my distaste when people fail in them (including myself), especially since they mostly amount to "don't be a dick, and if you are going to be a dick or accuse someone of something, have and show evidence."

From OP's point of view he paid for git tower, sent in criticism and them immediately after couldn't use it anymore because his license was revoked. That's his point of view.

The actual git tower folks as a PR measure came and said that 'because we couldnt figure out how our ticket system works the license was revoked and there was a bug.'

That seems rather far fetched.

The OP's point of view is valid. The OP's assertion as to the cause of their license being revoked was not backed up with any useful evidence (a coincidence is not sufficient). Here's some things the OP could have done, or included if they did do or know, to back up their claim:

- Explain how the Reddit post could have been linked to the Git Tower license, so even if they wanted to revoke the license, they would know how to.

- Explain what Git Tower said when asked why the license was revoked, or whether they were even asked at all.

- Note whether there could have or has previously been any oddness or confusion with licenses (in a later comment, OP notes "I tried to lookup my old license and realized I was given 1 free license this annual period (around ~October) after purchasing 2 seats the previous year for contributing significantly to bug reports with full screencasts, etc.")

In fact, if you check the reddit post now, you'll see the top comment is actually a post-mortem from the company explaining what happened. You can choose to still disbelieve what they say (a bug in the licensing system, they wouldn't do this, they didn't even know the account) if you like since you seem to think it's far fetched, but I think by far the most likely situation (even before I noted any comments from the developer) was that there was some bug or error, given the OP didn't seem to put any effort into figuring out what went on before making a very negative assumption and then posting publicly like it was true.

He addressed this in a comment: > I tried to lookup my old license and realized I was given 1 free license this annual period (around ~October) after purchasing 2 seats the previous year for contributing significantly to bug reports with full screencasts, etc.

Seems his license was renewed in October and was revoked just now, he couldn't have requested a refund because his license was free...

Julian here from Tower. We posted an updated response: https://www.reddit.com/r/git/comments/ao1q7t/care_git_tower_...

I used to love Git Tower until the most recent version came out with the annual-renewing license. I used to promote Git Tower all the time because the one-time-purchase license was such a good deal for the tool and I truly believed it was worth the cost. However, this is another strike against the tool. I already abandoned Git Tower and switched to using Magit in Emacs. I don't regret my decision.

A big part of the problem (why a lot of apps are doing this) is Apple disallows software vendors from charging for major release upgrades on the Mac App Store. Ideally Apple would provide a pathway for upgrade pricing as a one-time purchase but of course they have a perverse incentive to sell subscriptions, and developers take the blame for it.

I have no issue with paying for subscriptions or paying for major releases. For example, JetBrains products are subscription-based but after a year of paying, you gain a perpetual license and can continue using the product without continuing to pay for updates. The problem with Git Tower now is that you always have to have a subscription going to use the product. For me, the value of the product does not outweigh the cost.

Edit: grammar

One can implement the pricing model used by Sketch, and do it via In-App purchases. Pay and get updates for a year. After that no new features work in your copy or any updates to your copy, unless you pay again.

https://www.sketchapp.com/pricing/

As a software developer you should be more empathetic towards the author.

Doesn't he, like you, deserve a paycheck every month?

Of course.

But the reality is this: The cost has increased; and some customers are saying the new higher cost is pricing them out of the product.

The loss of those customers for higher profits on the remainder may make financial sense. But that is only known to the developer/author after a full assessment of the impact.

Many other companies use price discrimination to combat this (e.g. Personal licenses are cheaper than business licenses, perpetual usage w/o updates, free OSS/student licenses, etc). Jetbrains utilizes almost all those techniques.

Absolutely. But with every business, customers are going to churn whenever there's a change in billing model. For me, the yearly subscription price is not worth it.

Git tower speaks: https://www.reddit.com/r/git/comments/ao1q7t/care_git_tower_...

Even if revoking the license was intentional retaliation, how would the developers of Git Tower know who posted on Reddit (and therefore know who's license to revoke)?

I don't see anything giving them that right in https://www.git-tower.com/legal/eula, not a lawyer at all but I think OP has a case?

From the user in question:

> I tried to lookup my old license and realized I was given 1 free license this annual period (around ~October) after purchasing 2 seats the previous year for contributing significantly to bug reports with full screencasts, etc.

> So it looks like I received it for free (well, more or less - after roughly a combined 20 hours worth of bug report effort for the year).

Revoking a free license seems a little less shady than revoking a paid one.

"fournova reserves all rights not expressly granted to you within this Agreement" plausibly includes "revoking it because we don't like you" but you'd need to be a German contract lawyer to know what they'd make of it, since apparently it's governed by German law...

Pretty slimy. Even if the user had said something really nasty on reddit about the product revoking their license is a pretty ridiculous and petty thing to do.

If we really have all the facts here it seems like it is reason enough not to ever bother with Git Tower.

That seems like a pretty big "if" to be fair.

https://www.techdirt.com/articles/20161220/12411836320/softw...

You'd think they'd take a page...

Honestly, this doesn't seem very likely—if the Tower folks were assholes, I could see them try it after a very nasty review, but it wasn't a nasty review at all, so it seems unlikely Tower would try to retaliate for it.

Off topic: but is there anything for life-long git CLI user to gain from by going GUI?

If you're comfortable where you are, then no. Personally, I use a mixture of both. For browsing through commit history or viewing diffs before committing / pushing, I find the GUI to be much more convenient. For most other things, I use the command line.

- easier to pick partial commits - easy visualisation of where branches split and merge - same easy visualisation of stashes

I use both the CLI and GUI. They both have their merits and demerits. While a CLI may have faster commands, a GUI helps a lot when trying to make sense of relationships between commits branches and visualisation in general.

The main benefit for me is the ability to quickly and conveniently see the commit history and diff as well as branches' current position. If you're already comfortable with doing the same in CLI, then you're probably not going to gain a ton.

I like Sourcetree for this particularly because its buttons map fairly well to the actual CLI commands that it will be executing, and makes it painless to open a CLI terminal for instances where I want more fine-tuned control.

As someone who made the opposite transition: no. My colleagues that are still GUI bound are generally slower at accomplishing most day to day tasks. Committing, rebasing, merging, pushing tags, etc.

I wonder, how is that even possible. Like committing you have the same or less amount of work to do. Merging is just drag-and-drop. Rebasing is two clicks.

You really underestimate how fast CLI is.

For me the value is in diffing, merging and browsing the commit graph. Piecemeal staging can be improved looking at the whole file instead of the window `add -p` shows.

But... I'm still primarily a CLI user. I've just configured difftool and mergetool to call up Beyond Compare. git gui and gitk make designers weep, but I've never felt hindered or slowed down in what I want to accomplish with them.

I'll try a new GUI tool every now and then, and I haven't found any that make me feel more productive than the CLI augmented by the tools above.

I've found graph views and cherry picking to be better experiences in various guis, but generally for graphs I use tig or gitk and for cherry pick I suffer and copy paste commit hashes. The weight of the various guis and the obscurity of internals is usually enough to stop me from using it.

The only use I have found for a GUI is partial commits, and for that I use gitk which is already distributed with git.

Reply on news.ycombinator.com