Re:SlideMail – an email app that can think
by vu0tran on Tuesday February 17, 2015 @01:00PM
(#9063842)
Hey guys, Vu, the founder here.
I'm a developer and my email habits are absolutely horrible. Whenever I put in long hours of coding, I feel like the time it takes to context switch between development and doing email is way too high.
Because of this, I'm notorious for missing emails, flights even or just not responding to emails as it gets lost in my inbox forever.
I essentially started making this email app for myself and it's finally working quite well. There have been times where I've completely forgotten about a meeting and it's reminded me to show up on time. The sorting uses a lot of NLP magic and I find it to be pretty accurate too for me to take back control of my inbox. I just want to see how far I can run with it and already working on trying to suggest replies to emails.
There are a lot of email apps on the market and I think their concept is "how do we incorporate better design and work flow to solve the email problem"? My notion is rather, "how do we use technology to solve the email problem?"
Other than that, I'm also really excited with the privacy route I decided to take. The downside is that deploying across multiple platforms will be more challenging without a central server side component, but if the plus is that if one chooses the option of going "private" (essentially this turns off server side push notifications) there won't be any communication with my server at all.
It's been a real struggle these past few months so I'm super excited!!
Re:SlideMail – an email app that can think
by e1ven on Tuesday February 17, 2015 @01:50PM
(#9064173)
I wanted to thank you for doing the processing on-device.
Too many mail clients are trying to sync you mail up to their servers, which creates no end of problems - Security, Compliance, and just general creepiness.
I look forward to learning more about your app! I'll certainly try it out when it launches ;)
Re:SlideMail – an email app that can think
by amirmc on Tuesday February 17, 2015 @03:17PM
(#9064675)
Indeed! It's not just about privacy either. The less the reliance on someone else's servers the less likely that a subsequent shutdown is going to impact you (whether by acquisition or otherwise).
FWIW I'd like to see approaches like this applied to personal servers so that the device doesn't have to do all the heavy lifting but the code still runs close to the user. The approach I'm taking is based on unikernels http://nymote.org/blog/2013/introducing-nymote/
Re:SlideMail – an email app that can think
by grinich on Tuesday February 17, 2015 @03:09PM
(#9064631)
Can you expand on this a bit? I work at Nilas, which provides sync for email, contacts, and calendar sync, along with modern REST APIs. https://nilas.com/sync_engine
We've had a lot of developers choose to use our infrastructure instead of interfacing with the various different providers and protocols (include Exchange). Usually security/compliance isn't an issue, but I'm looking to learn more about where folks have concerns, and ways we can address them. (Potentially doing everything from FHE to shipping AMIs.)
Re:SlideMail – an email app that can think
by __mp on Tuesday February 17, 2015 @04:07PM
(#9064993)
I currently have 20GB of email lying around on several mail services. There is no way I want to allow an additional app to my personal mail account for email management. The fact that I'm hosting my mail somewhere else is bad enough. I'm paying money for mail hosting since I want the person running the service to keep my data safe and confidential.
I'm currently happy directly accessing my email via web, Mail.app and Maildroid. Using a mail processing provider would mean that I would need to read another TOS to make sure my data is not sold/processed/used. If a mail client isn't able to speak standard imap and process the data locally then I don't even want to use it.
Maybe you should license your imap/api code for these thirdparty developers.
Re:SlideMail – an email app that can think
by urda on Tuesday February 17, 2015 @04:31PM
(#9065169)
Unless you're PGP'ing everything that comes and goes from your mailbox, you have Z E R O guaranteed security. Sure your mail server host may claim that they are hosting only, but you can never be sure that an employee has not gone rouge. What if the host decides to start monitoring your mail server upstream?
Host the server at home instead? Well do you trust your ISP? I imagine you don't.
Trust along the entire transit of the message, as well as the final destination? Can't there either.
So in the end it doesn't matter if there's yet another service in the middle. API or host or anything. If you're not encrypting it locally it doesn't matter. Period. The only point you can sell on is you are removing a service that can cause a point of failure. But don't act like it's more secure than before.
Re:SlideMail – an email app that can think
by __mp on Wednesday February 18, 2015 @04:21PM
(#9071031)
I'm well aware of that. Right now it's just as important for me to try to manage my security and privacy.
Hosting the mail server at home would be a possibility (we have a fiber connection). I could also place a server in the datacenter of a friend and do the mail hosting myself. I don't want to spend the time to manage my email hosting since I'm depending on it. Running your own mailserver is not as easy as it sounds like. It's rather time consuming when you want to do it in a sound way:
- Spam filtering
- DKIM
- SPF
- Security
etc... Sure I could follow a $random_tutorial but then I would surely miss a couple of best practices. I don't think this time is well spent because I'm paying other people more knowledgeable about email to do it for me. That's a conscious decision I made. I rather not have an additional party involved when I see no need.
Please note that also used to run/help run a couple of large mail servers. So I have a general idea how they behave and what to do.
Re:SlideMail – an email app that can think
by urda on Tuesday February 17, 2015 @06:37PM
(#9065959)
But that's how security works. Reducing the surface area is great, nothing wrong with that. However, the system is only as strong as the weakest link. E-mail is hilariously insecure, so just avoiding a single layer, API, or applications and calling it done is not enough.
Re:SlideMail – an email app that can think
by Mithaldu on Tuesday February 17, 2015 @01:07PM
(#9063892)
I can't tell for what OS this is, and unless you're planning to make it for both android and ios, please put the fact that it's only for one, and which it is, top and large on your site.
There is little that makes me think worse of a company than having it breathlessly describe a product i actually want, without telling me whether i can actually use it.
Re:SlideMail – an email app that can think
by mynameisvlad on Tuesday February 17, 2015 @02:31PM
(#9064433)
Literally every single product image uses an iPhone, and the real product photo is also clearly an iPhone.
Just because it doesn't have a big image or text with OS compatibility does not mean you can't intelligently determine based on the product photos what OS it's for. If it were for Android, they'd have at least one Android photo, or specifically mention Android compatibility. If it's a completely iOS-centric page, you can guess /quite easily/ it's iOS only.
Re:SlideMail – an email app that can think
by Mithaldu on Tuesday February 17, 2015 @06:53PM
(#9066045)
I don't know anyone with an iphone and rarely see pictures of them. As such, i simply am not familiar enough with them to figure out whether it's an android or iphone just by looking at it, especially since the UI is very generic and android mods and ISP OSes plentiful.
Re:SlideMail – an email app that can think
by halflings on Tuesday February 17, 2015 @02:11PM
(#9064317)
I really do hope there's an android version planned.
It'd be great to have some competition in the field of "intelligent email clients". Inbox is doing a great job, but it's for Gmail only.
Re:SlideMail – an email app that can think
by 51Cards on Tuesday February 17, 2015 @02:10PM
(#9064309)
True but not uncommon to just use one device for the marketing. Also went looking for the platform icons myself as I would have tried this out on Android.
Re:SlideMail – an email app that can think
by pbreit on Tuesday February 17, 2015 @04:31PM
(#9065172)
The main thing I'm looking for is automatic segregation in between Outlook and Gmail. I'd like 1 more bucket in Outlook so that other splits between sorta important and not important. Gmail segregates by type which is not always the best proxy for importance.
I'd like a Focused bucket which is senders I know, replies to me emails and otherwise important stuff (important enough that I want the iOS notification). The middle bucket is stuff I occasionally open. The third bucket is stuff I haven't unsubscribed to yet for whatever reason.
Re:SlideMail – an email app that can think
by ilyaeck on Tuesday February 17, 2015 @05:28PM
(#9065559)
You seem to be trying to resolve 2 different pains at once: email overload and privacy. Regarding the former, Google seems to be doing a lot of the same stuff already (categorization, calendar event extraction, ...). Do you have your unique edge? Or is the differentiation mainly in not being tied to Gmail?
Re:SlideMail – an email app that can think
by spost on Tuesday February 17, 2015 @06:22PM
(#9065880)
Not OP, but I look at this not as trying to solve email overload and privacy, but as trying to solve email overload with the nifty side benefit of being more privacy-conscious than many of the alternatives.
Re:SlideMail – an email app that can think
by jlees on Tuesday February 17, 2015 @01:38PM
(#9064094)
Would love to hear more about the NLP approaches you've been using, and how well they've generalized to other folks. Though that may be giving away too much of the secret sauce, of course.
Re:SlideMail – an email app that can think
by hayksaakian on Tuesday February 17, 2015 @04:18PM
(#9065085)
Three Practical tips to get your slidemail app more sign ups:
- You should have a call to action at the bottom of the page. It would be as simple as copy/pasting the code for the one-line email capture from above
- encourage sharing. i didn't provide my email address to test this, but you should encourage people (at some point: before signing up, right after signing up, or in a follow up email) to share this thing.
- I got the impression that it's iOS only. You probably got a lot of visits from android devices (check your own logs if you don't believe me). You could address that with a separate capture form using User Agent detection? Something like: "Get notified when we launch the Android version"
small nit pick about the page:
- <meta> tags are like <img> tags they should NOT have a corresponding close tag, "</meta>"
good luck with the app Vu
(i'm still using your old "hack" parakeet so i know you make some interesting stuff)
Re:SlideMail – an email app that can think
by maelito on Tuesday February 17, 2015 @12:49PM
(#9063746)
"All the processing happens within the SlideMail app itself. Whether you're syncing through Gmail, Yahoo Mail, AOL Mail, or any other IMAP based email service, your emails will never be stored outside of your app. If you delete the application from your phone, your data is gone for good."
This is interesting.
Re:SlideMail – an email app that can think
by vu0tran on Tuesday February 17, 2015 @01:02PM
(#9063859)
Yeah, totally. I think with these features, one group that it has resonated with that was unexpected were college students and people that are less tech saavy like my mom and my dad.
They still use Yahoo mail so Inbox / Gmail just doesn't work for them. SlideMail sort of brings what the Gmail app has to offer, in my opinion, enhances it and provides it for IMAP in general as well as other services.
We also did the "only check mail on the phone" implementation for security, but it ends up hurting the user experience because Apple gets to decide when to use background data.
It gets better (more frequent) the more you use the app, but it's hard to get users over the initial delays in mail checking. Just something to think about as you're building this.
Re:SlideMail – an email app that can think
by tthayer on Tuesday February 17, 2015 @01:13PM
(#9063932)
Does it support IMAP IDLE? I've had to do some ridiculous workaround by forwarding my work and gmail accounts to icloud accounts just so I can get push working and none of the other email clients I have tried support it.
Re:SlideMail – an email app that can think
by wingerlang on Wednesday February 18, 2015 @06:30AM
(#9067954)
I don't like that the first thing I see is an "sign up now" and the entering of my email. Show me what you've got before trying to get (Only the impression) my email.
Re:SlideMail – an email app that can think
by Killswitch on Wednesday February 18, 2015 @02:46PM
(#9070538)
Well you have two options, pay for email, or use gmail for free in exchange for allowing this.
I'm all for privacy, but sometimes you guys go way overboard and complain and act entitled like Google is obligated to offer Gmail free of charge with nothing in return.
Re:SlideMail – an email app that can think
by ocdtrekkie on Wednesday February 18, 2015 @04:36PM
(#9071115)
I'd be happy to pay for good private email service. But of course, Google doesn't offer it. Because offering that sort of option would be a much more obvious admission of how much they rifle through your data as an income source.
My issue is not that they offer a business relationship where they use your data for money, my issue is how shady they are about it.
Re:SlideMail – an email app that can think
by Killswitch on Wednesday February 18, 2015 @06:34PM
(#9071606)
Google has always been obvious about all their products and services being for selling more advertising.. They figured out how to make a great income from it, and nobody would pay for email at the cost it'd have to be to make the same as they do.
Remember, all the good stuff they do, all depends on revenue dollars coming in. And in the end, they're still a company looking to make a profit.
Re:SlideMail – an email app that can think
by Animats on Tuesday February 17, 2015 @06:05PM
(#9065790)
Agreed. This provides most of the functionality of GMail without the ads and snooping.
Try advertising this to lawyers, who can't use GMail because Google snoops the mail. Consider offering a premium version that understands more legal terminology about things like court dates.
Re:SlideMail – an email app that can think
by dang on Tuesday February 17, 2015 @08:09PM
(#9066410)
You're right. Show HN is for things that can be tried out now. It's ok if people have to download an app to do that. But an email signup to get "early invites by May" does not count—it's the very thing that rule was written to exclude.
If your work isn't ready for people to try out yet, please don't put "Show HN" in the title. Once it's ready, come back and share it then.https://news.ycombinator.com/showhn.html
It's good for the community to tell us about these, since we can't notice them all. In the future, you (or anyone) are welcome to email us at [email protected] to make sure we look at it.
Re:SlideMail – an email app that can think
by veb on Tuesday February 17, 2015 @04:01PM
(#9064937)
The difference is... they offer nothing except "we made something, sign up?!" whereas this is actually a lot different. If you read the thread, the founder is in here commenting and you can tell that there's already a working product. Polished and shiny it may not be, but this is not just empty space like most LaunchRock type pages.
Re:SlideMail – an email app that can think
by vishalzone2002 on Tuesday February 17, 2015 @04:05PM
(#9064975)
First difference doesn't even count. With no offense to the founder, its easy to take part in the discussion.
How can you tell that there is actually a working prototype/MVP? These wireframes can be made in proto or PSDs.
Re:SlideMail – an email app that can think
by prawn on Tuesday February 17, 2015 @05:19PM
(#9065500)
How is it a lot different? If I click the misleading "Get app" button, it shunts me down to the email capture which says that even the early subscribers won't get anything for three months.
I don't think the fact that the founder is commenting is relevant. I could chat with people about a theoretical or in-progress product quite easily.
I've flagged this submission. However much work has gone into it, GP is correct in saying that it's an email sign-up. Better for OP to wait until there is a product for people to use.
Re:SlideMail – an email app that can think
by Kiro on Wednesday February 18, 2015 @03:17AM
(#9067510)
What's even worse are the people posting projects that are not their own as Show HNs. I see it all the time. Some people just think Show HN means "look what I found on the internet".
Re:SlideMail – an email app that can think
by djrogers on Tuesday February 17, 2015 @04:10PM
(#9065029)
>> Show HN is for something you've made that other people can play with. HN users can try it out, give you feedback, and ask questions in the thread.
>Which is what it is. It's not just an email sign-up. I think this is a perfect example of a good Show HN.
Until there is more than just a website with an email signup (i.e., an actual shipping email client), then it isn't something HN users can try out or give feedback on.
Re:SlideMail – an email app that can think
by veb on Wednesday February 18, 2015 @06:26AM
(#9067945)
Thank you for your explanation. I thought people could give feedback based on what was on the site, ask the author/creator questions/feedback in the thread. I did say it was my opinion that I thought it was a good example, but we've all got different ones!
Again, thanks for actually bothering to make an effort to let me know -- rather than the people who downvted me without bothering to explain anything. Thanks mate.
I'm a developer and my email habits are absolutely horrible. Whenever I put in long hours of coding, I feel like the time it takes to context switch between development and doing email is way too high.
Because of this, I'm notorious for missing emails, flights even or just not responding to emails as it gets lost in my inbox forever.
I essentially started making this email app for myself and it's finally working quite well. There have been times where I've completely forgotten about a meeting and it's reminded me to show up on time. The sorting uses a lot of NLP magic and I find it to be pretty accurate too for me to take back control of my inbox. I just want to see how far I can run with it and already working on trying to suggest replies to emails.
There are a lot of email apps on the market and I think their concept is "how do we incorporate better design and work flow to solve the email problem"? My notion is rather, "how do we use technology to solve the email problem?"
Other than that, I'm also really excited with the privacy route I decided to take. The downside is that deploying across multiple platforms will be more challenging without a central server side component, but if the plus is that if one chooses the option of going "private" (essentially this turns off server side push notifications) there won't be any communication with my server at all.
It's been a real struggle these past few months so I'm super excited!!
Too many mail clients are trying to sync you mail up to their servers, which creates no end of problems - Security, Compliance, and just general creepiness.
I look forward to learning more about your app! I'll certainly try it out when it launches ;)
FWIW I'd like to see approaches like this applied to personal servers so that the device doesn't have to do all the heavy lifting but the code still runs close to the user. The approach I'm taking is based on unikernels http://nymote.org/blog/2013/introducing-nymote/
We've had a lot of developers choose to use our infrastructure instead of interfacing with the various different providers and protocols (include Exchange). Usually security/compliance isn't an issue, but I'm looking to learn more about where folks have concerns, and ways we can address them. (Potentially doing everything from FHE to shipping AMIs.)
Thanks! You can also ping me directly at [email protected].
I'm currently happy directly accessing my email via web, Mail.app and Maildroid. Using a mail processing provider would mean that I would need to read another TOS to make sure my data is not sold/processed/used. If a mail client isn't able to speak standard imap and process the data locally then I don't even want to use it.
Maybe you should license your imap/api code for these thirdparty developers.
Host the server at home instead? Well do you trust your ISP? I imagine you don't.
Trust along the entire transit of the message, as well as the final destination? Can't there either.
So in the end it doesn't matter if there's yet another service in the middle. API or host or anything. If you're not encrypting it locally it doesn't matter. Period. The only point you can sell on is you are removing a service that can cause a point of failure. But don't act like it's more secure than before.
Hosting the mail server at home would be a possibility (we have a fiber connection). I could also place a server in the datacenter of a friend and do the mail hosting myself. I don't want to spend the time to manage my email hosting since I'm depending on it. Running your own mailserver is not as easy as it sounds like. It's rather time consuming when you want to do it in a sound way: - Spam filtering - DKIM - SPF - Security etc... Sure I could follow a $random_tutorial but then I would surely miss a couple of best practices. I don't think this time is well spent because I'm paying other people more knowledgeable about email to do it for me. That's a conscious decision I made. I rather not have an additional party involved when I see no need.
Please note that also used to run/help run a couple of large mail servers. So I have a general idea how they behave and what to do.
(And don't bother to PGP encrypt them as you can't trust the Debian GnuPG maintainer either.)
Where the heck was that written? That's not how Public / Private Keys work AT ALL.
> (And don't bother to PGP encrypt them as you can't trust the Debian GnuPG maintainer either.)
You do realize you can download, review, and compile from source? It's called Open Source for that very reason.
I expect more of an educated response on Hacker News than the one you have just demonstrated here.
There is little that makes me think worse of a company than having it breathlessly describe a product i actually want, without telling me whether i can actually use it.
Just because it doesn't have a big image or text with OS compatibility does not mean you can't intelligently determine based on the product photos what OS it's for. If it were for Android, they'd have at least one Android photo, or specifically mention Android compatibility. If it's a completely iOS-centric page, you can guess /quite easily/ it's iOS only.
iOS 7 (8 is pretty much the same): http://now.avg.com/wp-content/uploads/2013/09/ios7_homescree...
Android Lollipop: http://upload.wikimedia.org/wikipedia/commons/5/52/Android_5...
Furthermore, I doubt they'd do anything crazy if they were to show an Android app. It's usually stock or very few mods.
I'd like a Focused bucket which is senders I know, replies to me emails and otherwise important stuff (important enough that I want the iOS notification). The middle bucket is stuff I occasionally open. The third bucket is stuff I haven't unsubscribed to yet for whatever reason.
- You should have a call to action at the bottom of the page. It would be as simple as copy/pasting the code for the one-line email capture from above
- encourage sharing. i didn't provide my email address to test this, but you should encourage people (at some point: before signing up, right after signing up, or in a follow up email) to share this thing.
- I got the impression that it's iOS only. You probably got a lot of visits from android devices (check your own logs if you don't believe me). You could address that with a separate capture form using User Agent detection? Something like: "Get notified when we launch the Android version"
small nit pick about the page:
- <meta> tags are like <img> tags they should NOT have a corresponding close tag, "</meta>"
good luck with the app Vu
(i'm still using your old "hack" parakeet so i know you make some interesting stuff)
~ @hayksaakian on twitter
They still use Yahoo mail so Inbox / Gmail just doesn't work for them. SlideMail sort of brings what the Gmail app has to offer, in my opinion, enhances it and provides it for IMAP in general as well as other services.
We also did the "only check mail on the phone" implementation for security, but it ends up hurting the user experience because Apple gets to decide when to use background data.
It gets better (more frequent) the more you use the app, but it's hard to get users over the initial delays in mail checking. Just something to think about as you're building this.
Gmail / iCloud: Yes FastMail: Yes Yahoo: No AoL / Exchange / Hotmail / Live: Can't remember
Kudos to this. Bunch of scammers like Google try convincing you they HAVE to have your data for these things to be possible.
I'm all for privacy, but sometimes you guys go way overboard and complain and act entitled like Google is obligated to offer Gmail free of charge with nothing in return.
My issue is not that they offer a business relationship where they use your data for money, my issue is how shady they are about it.
Remember, all the good stuff they do, all depends on revenue dollars coming in. And in the end, they're still a company looking to make a profit.
Try advertising this to lawyers, who can't use GMail because Google snoops the mail. Consider offering a premium version that understands more legal terminology about things like court dates.
"... blog posts, email signups, and fundraisers can't be tried out, so they don't count as Show HNs" (https://news.ycombinator.com/showhn.html)
If your work isn't ready for people to try out yet, please don't put "Show HN" in the title. Once it's ready, come back and share it then. https://news.ycombinator.com/showhn.html
It's good for the community to tell us about these, since we can't notice them all. In the future, you (or anyone) are welcome to email us at [email protected] to make sure we look at it.
I don't think the fact that the founder is commenting is relevant. I could chat with people about a theoretical or in-progress product quite easily.
I've flagged this submission. However much work has gone into it, GP is correct in saying that it's an email sign-up. Better for OP to wait until there is a product for people to use.
> Show HN is for something you've made that other people can play with. HN users can try it out, give you feedback, and ask questions in the thread.
Which is what it is. It's not just an email sign-up. I think this is a perfect example of a good Show HN.
>Which is what it is. It's not just an email sign-up. I think this is a perfect example of a good Show HN.
Until there is more than just a website with an email signup (i.e., an actual shipping email client), then it isn't something HN users can try out or give feedback on.
Again, thanks for actually bothering to make an effort to let me know -- rather than the people who downvted me without bothering to explain anything. Thanks mate.
Edit: or is the intent here that I will get immediate access if I put in my email? And the 'invites' is for the ready product?