1
00:00:00,000 --> 00:00:08,760
Welcome to the NGI Zero podcast where we talk to the people who are building the Next Generation

2
00:00:08,760 --> 00:00:09,760
Internet.

3
00:00:09,760 --> 00:00:12,080
Hi, I'm Ronny Lam.

4
00:00:12,080 --> 00:00:13,940
And I'm Tessel Renzenbrink.

5
00:00:13,940 --> 00:00:19,440
We both work for NLnet, a foundation which financially supports people working on free

6
00:00:19,440 --> 00:00:21,760
and open source technologies.

7
00:00:21,760 --> 00:00:23,680
Our guest today is Adrian Georgescu.

8
00:00:23,680 --> 00:00:31,080
He is the founder and CEO of AG Projects, a company committed to open source telecommunications.

9
00:00:31,080 --> 00:00:36,200
For many years Adrian has worked on Sylk Suite, an open source multi-party conference tool.

10
00:00:36,200 --> 00:00:38,640
And before Sylk, there was Blink.

11
00:00:38,640 --> 00:00:41,920
We'll be talking about both these projects today.

12
00:00:41,920 --> 00:00:44,840
Hi Adrian, nice to have you here.

13
00:00:44,840 --> 00:00:46,680
Hello Tessel, hello Ronny.

14
00:00:46,680 --> 00:00:50,040
Thank you for inviting me to this session.

15
00:00:50,040 --> 00:00:56,080
We have devised a list of questions to get to know our guests quickly.

16
00:00:56,080 --> 00:00:59,840
They are three short questions.

17
00:00:59,840 --> 00:01:05,080
So Firefox or Brave web browser?

18
00:01:05,080 --> 00:01:09,520
This is like voting left or right, the outcome is the same.

19
00:01:09,520 --> 00:01:13,120
You didn't answer yet.

20
00:01:13,120 --> 00:01:14,120
Firefox.

21
00:01:14,120 --> 00:01:19,940
Okay, vampires or zombies?

22
00:01:19,940 --> 00:01:24,820
This is the type of, I have a third option, none of them.

23
00:01:24,820 --> 00:01:28,380
Okay, telephone or telegraph?

24
00:01:28,380 --> 00:01:29,620
Blink.

25
00:01:29,620 --> 00:01:32,460
Good, yeah.

26
00:01:32,460 --> 00:01:36,740
So you work on communication systems.

27
00:01:36,740 --> 00:01:41,480
What key issues do you see in this field?

28
00:01:41,480 --> 00:01:50,660
This field has been spawned while internet expanded beyond its original goal, I guess.

29
00:01:50,660 --> 00:01:56,960
There was what we call in our niche the VoIP revolution.

30
00:01:56,960 --> 00:02:02,600
This happened a while ago when the telecommunications market has been liberalized in Western Europe

31
00:02:02,600 --> 00:02:04,660
and beyond.

32
00:02:05,420 --> 00:02:12,020
Internet connections became fast enough to accommodate real-time communications and that's

33
00:02:12,020 --> 00:02:25,380
when internet started to grow beyond the e-commerce and email and messaging systems to what we

34
00:02:25,380 --> 00:02:29,180
have today, these video conferencing applications.

35
00:02:29,180 --> 00:02:36,060
Actually it was Voice over IP, the first real-time application that emerged to be ported from

36
00:02:36,060 --> 00:02:43,500
the old telecom network to the internet and this VoIP revolution, I was part of it and

37
00:02:43,500 --> 00:02:53,400
it helped me discover my own success by riding this wave.

38
00:02:53,400 --> 00:02:57,140
So I was early and I had the luck to have a good mentor.

39
00:02:57,140 --> 00:03:03,580
That is Henry Sinnreich, who was perceived in the community as the godfather of Voice

40
00:03:03,580 --> 00:03:10,620
over IP because he was instrumental in funding the first open source projects related to

41
00:03:10,620 --> 00:03:13,980
Voice over IP worldwide.

42
00:03:13,980 --> 00:03:20,500
I got to know this part of the industry, the emerging industry, very early on and I was

43
00:03:20,500 --> 00:03:30,860
part of those first creators that made Voice over IP work based on internet principles,

44
00:03:30,860 --> 00:03:39,060
which is something I would like to mention because internet was built based on certain

45
00:03:39,060 --> 00:03:48,560
principles, one of them being the end-to-end principle, which means that the network should

46
00:03:48,600 --> 00:03:57,080
be kept as simple as possible and servers should be doing the least amount of work possible

47
00:03:57,080 --> 00:04:03,720
and most of the work and the application should be developed in endpoints, the network being

48
00:04:03,720 --> 00:04:09,120
a transparent carrier for whatever the applications in the endpoints are meant or supposed to

49
00:04:09,120 --> 00:04:10,720
do.

50
00:04:10,720 --> 00:04:15,640
And this end-to-end principle was embedded into the products of mine and other companies

51
00:04:15,720 --> 00:04:20,480
who stuck to this principle as opposed to others who tried to do the opposite, putting

52
00:04:20,480 --> 00:04:26,840
a lot of functionality in the network and letting the end user with very little choices

53
00:04:26,840 --> 00:04:31,720
and no transparency to what happens in the central part of the internet.

54
00:04:31,720 --> 00:04:39,240
So how does Sylk contribute to addressing those issues?

55
00:04:39,240 --> 00:04:47,240
When the end-to-end principle was such that email is the best example of a software that

56
00:04:47,240 --> 00:04:52,480
its principles are today, you can verify that the principles applied during email are still

57
00:04:52,480 --> 00:04:57,540
valid today because we're still using emails after so many years.

58
00:04:57,540 --> 00:05:05,240
When Voice over IP became an issue, the way to make it interoperable and making a standard

59
00:05:05,240 --> 00:05:09,240
so that everyone can implement solutions that talk to each other.

60
00:05:09,240 --> 00:05:18,880
The IETF, the Internet Engineering Task Force, created a working group which eventually led

61
00:05:18,880 --> 00:05:24,800
to the creation of this protocol called SIP, which stands for Session Initiation Protocol,

62
00:05:24,800 --> 00:05:30,400
which was basically copying from the success of email and HTTP.

63
00:05:30,400 --> 00:05:36,540
It is a simple protocol to develop around.

64
00:05:36,540 --> 00:05:41,960
It could be reusing the existing email and HTTP models.

65
00:05:41,960 --> 00:05:45,160
It was a predictable success and it is a success.

66
00:05:45,160 --> 00:05:50,120
SIP protocol is being embedded into any telecommunication backbone there is in the world.

67
00:05:50,120 --> 00:05:55,080
Any cable operator uses SIP protocol at the base, though they don't advertise it as such.

68
00:05:55,080 --> 00:06:01,560
The protocol was and still is a success, but less for the end users who don't get the immediate

69
00:06:01,560 --> 00:06:07,080
benefits of using SIP because they are being offered services by companies like cable operators,

70
00:06:07,080 --> 00:06:13,680
fiber providers, and they are given an address typically in the form of a phone number.

71
00:06:13,680 --> 00:06:17,880
They are not told that they are using SIP and they can't even make use of all the features

72
00:06:17,880 --> 00:06:22,920
of the SIP protocol because of the business aspect of using this.

73
00:06:22,920 --> 00:06:29,800
The companies just replaced telephony in place of letting SIP being used for what is originally

74
00:06:29,800 --> 00:06:35,800
designed for any type of session, including video, presence, chat, file transfers, any

75
00:06:35,800 --> 00:06:39,360
type of real-time communications between end users.

76
00:06:39,360 --> 00:06:46,040
Today we see still telephony being as the main application and our company, since we

77
00:06:46,520 --> 00:06:54,360
started, we were dedicated to make SIP a success the way the protocol is envisaged, meaning for

78
00:06:54,360 --> 00:07:02,600
applications beyond VoIP and we're the first company that being committed to that we really

79
00:07:02,600 --> 00:07:10,200
implemented most, if not all, the specifications from the IETF related to establishing sessions

80
00:07:10,200 --> 00:07:19,240
beyond VoIP, namely chat, messaging, file transfer, conferencing, video calls, and we

81
00:07:19,240 --> 00:07:24,520
had this very early on when nobody could even use them because of lack of deployments, lack

82
00:07:24,520 --> 00:07:26,120
of bandwidth, lack of everything.

83
00:07:26,120 --> 00:07:36,200
However, the software, when IETF proposes ideas and working groups are drafting plans and

84
00:07:36,200 --> 00:07:38,520
specifications and standards.

85
00:07:38,520 --> 00:07:46,440
It's always usable and useful to have an implementation to prove or disprove the attempted

86
00:07:46,440 --> 00:07:52,360
tracks and we were one of the companies who were instrumental in proving that some ideas

87
00:07:52,360 --> 00:07:59,720
were good and others were wrong, being involved close into the standardization and by means of

88
00:07:59,720 --> 00:08:04,200
creating software that implemented those specifications very early on while actually

89
00:08:04,200 --> 00:08:05,720
developing those specifications.

90
00:08:06,680 --> 00:08:12,280
Sylk is actually the second generation of software our company worked on.

91
00:08:12,280 --> 00:08:14,920
Initially it was Blink, which was a pure SIP client.

92
00:08:14,920 --> 00:08:16,120
There was SIP and nothing else.

93
00:08:17,400 --> 00:08:19,720
SIP protocol became complicated.

94
00:08:19,720 --> 00:08:24,360
Other things were necessary to do things that the protocol alone could not do.

95
00:08:24,360 --> 00:08:29,720
So other protocols emerged like there was an ecosystem around SIP and things were created

96
00:08:29,800 --> 00:08:37,640
around SIP to help SIP achieve the deliverables that were initially aimed for could not be

97
00:08:38,200 --> 00:08:44,120
achieved because there was more work required and more parts and other protocols were necessary

98
00:08:44,120 --> 00:08:45,560
to complete a picture.

99
00:08:45,560 --> 00:08:52,200
And Blink was a pure SIP client doing everything SIP and then the limitations of the SIP protocol

100
00:08:52,200 --> 00:08:55,640
started to appear while more complex applications started to emerge.

101
00:08:55,720 --> 00:09:00,600
For instance, this video conference we have right now practically is not possible to do

102
00:09:00,600 --> 00:09:07,400
in SIP because SIP has no primitives that can allow you to develop multi-party conferencing,

103
00:09:07,400 --> 00:09:08,680
not in a way that makes sense.

104
00:09:09,720 --> 00:09:18,360
So SIP stopped short at a point where video became successful and people started to adopt

105
00:09:19,720 --> 00:09:22,520
video calls, especially in video conferencing format.

106
00:09:23,240 --> 00:09:29,080
SIP stopped at end-to-end, peer-to-peer, one-to-one calls and beyond that,

107
00:09:30,040 --> 00:09:32,600
standardization didn't help but rather the opposite.

108
00:09:33,560 --> 00:09:38,760
So the end result was that there was no SIP implementation doing video conferencing,

109
00:09:38,760 --> 00:09:42,840
multi-party anywhere that made any sense at all technically and otherwise.

110
00:09:43,960 --> 00:09:50,920
And the alternative was provided by the web working groups which were not inside IETF but

111
00:09:51,000 --> 00:09:59,240
other people that leveraged the success of the web basically completed what SIP was missing,

112
00:09:59,240 --> 00:10:08,040
which is WebRTC, which added real-time media to web browsers and basically it became suddenly easy,

113
00:10:08,040 --> 00:10:13,160
not by standards but by adoption of certain pieces of code to make video conferencing possible

114
00:10:13,160 --> 00:10:18,760
in a web browser, which was not possible in a SIP protocol using the tools and the standards

115
00:10:18,760 --> 00:10:20,840
available were simply not doable.

116
00:10:21,560 --> 00:10:28,280
So Sylk came to join together these two worlds whereby SIP protocol can be used for what it does

117
00:10:28,280 --> 00:10:35,240
best, namely addressing in the form of an email address and using your own domain, you can have

118
00:10:35,240 --> 00:10:40,840
an identity which makes sense, a user@domain and then connect it to all the media that people

119
00:10:40,840 --> 00:10:48,120
may need, namely voice over IP, video, chat, but also the missing part, multi-party video

120
00:10:48,120 --> 00:10:50,440
conferencing, how do you do that with a SIP client?

121
00:10:51,480 --> 00:10:58,280
Well you can't, however, as I said, Sylk is supposed to be interoperable and for all the

122
00:10:58,280 --> 00:11:06,600
new features it's using WebRTC and for compatibility purposes it's using SIP protocols so you are

123
00:11:06,600 --> 00:11:12,120
always reachable in the best way that the client can support while negotiating a session with

124
00:11:12,120 --> 00:11:17,960
anyone that may call you on that address using SIP protocol and while using the same application

125
00:11:17,960 --> 00:11:24,520
like Sylk you can of course enjoy features particular to it, namely multi-party conferencing

126
00:11:24,520 --> 00:11:31,560
which is always, how do you say, locked to your, how do you say, not vendor but solution, there is

127
00:11:31,560 --> 00:11:37,080
no interoperable multi-party video conferencing solution that I know of so if you're on Zoom

128
00:11:37,080 --> 00:11:42,680
there is no way to join with another client, there is no way to, because there is no standard,

129
00:11:43,560 --> 00:11:48,920
standards became too complicated, too many people got involved and it ended up that video party

130
00:11:48,920 --> 00:11:55,880
multi-party conferencing it's not something standard... standardisable, so we ended up with

131
00:11:56,440 --> 00:12:02,440
multitude, a plethora of solutions to allow the users to do this and no standard way of

132
00:12:02,440 --> 00:12:10,440
interconnecting between them. Sylk is doing the same as it's not a weakness we could fix, it's

133
00:12:10,440 --> 00:12:16,680
inheriting the same model whereby you can start a video conference inside Sylk with other participants

134
00:12:17,320 --> 00:12:21,480
and then if you want to make a regular session one-to-one you can do it with anyone that supports

135
00:12:21,480 --> 00:12:28,440
SIP protocol worldwide with other solutions that implement SIP protocol and this is where they are

136
00:12:28,440 --> 00:12:36,200
alike or not alike the solutions. Blink is design-wise it's an application based on the

137
00:12:37,320 --> 00:12:44,840
you could say legacy paradigm. Web development changed the way interfaces look like a lot

138
00:12:45,480 --> 00:12:52,760
in the past you had different operating systems with user interfaces that looked in a certain way

139
00:12:52,760 --> 00:12:57,400
now you see a sort of standardization whereby all the application looked the same in a webish way

140
00:12:58,440 --> 00:13:06,360
in Blink it's a standard legacy type of software whereby we used standard desktop

141
00:13:06,360 --> 00:13:11,320
development and standard the widget set that is present on the natively on each computer

142
00:13:12,040 --> 00:13:19,080
so Blink is an interface that looks native on each desktop you use whether it's Linux or Mac

143
00:13:19,080 --> 00:13:25,640
or Windows whereas Sylk it's a cross-platform product based on these new principles of the web

144
00:13:25,640 --> 00:13:31,160
development so it looks the same everywhere but doesn't look native anywhere so this is the

145
00:13:32,760 --> 00:13:40,680
world we live in today. It's interesting to hear what you're saying is that the telephone

146
00:13:40,680 --> 00:13:49,560
companies are selling us telephone connections when actually they are going over SIP already

147
00:13:49,560 --> 00:13:56,920
for decades. 100% yes only SIP they don't advertise it but it's whenever you make a phone call from

148
00:13:56,920 --> 00:14:03,800
your telephone it's in the server side it's a SIP server and your phone is a SIP device restricted

149
00:14:03,880 --> 00:14:10,840
to do only on audio call. When I listen to your story I think you really made

150
00:14:12,280 --> 00:14:22,360
WebRTC and SIP work together they go yeah hand in hand so to say

151
00:14:25,320 --> 00:14:31,560
do you know any... are you the first are there any other applications that do it this well?

152
00:14:33,800 --> 00:14:41,080
With Blink definitely there was no we were the best and still are the I don't know about any

153
00:14:41,080 --> 00:14:46,440
other SIP client that aimed so high and achieved so much especially with the help of NLnet and

154
00:14:46,440 --> 00:14:55,880
other sponsors in the past. The real problem in my field of work meaning if you want to have a

155
00:14:55,880 --> 00:15:01,000
business and live from the fruits of your labor it's very hard because you compete with

156
00:15:01,400 --> 00:15:07,160
big companies that offer a free service that does what your software does but the question comes

157
00:15:07,160 --> 00:15:12,280
how do you live if Google gives you an alternative to make VoIP calls and Apple does the same and

158
00:15:12,840 --> 00:15:16,840
WhatsApp and all these companies that have thousands of developers and gives you something

159
00:15:16,840 --> 00:15:22,280
for free I'm not even sure why is there an expectation that any small company can be

160
00:15:22,280 --> 00:15:29,960
successful of doing the same with a few people in terms of revenues or success as a business

161
00:15:31,000 --> 00:15:38,760
I don't know how we wouldn't be successful had we just tried to monetize this SIP clients or

162
00:15:38,760 --> 00:15:43,240
Sylk clients because you compete with big names that offer a similar product which

163
00:15:44,840 --> 00:15:50,360
is properly maintained over years they have a lot of engineering support for it so it's a lost

164
00:15:50,360 --> 00:15:55,960
battle in the sense of trying to make money with something which is free. However knowing this

165
00:15:55,960 --> 00:16:02,440
again my company was able to position itself in a way that was able to maintain over time competing

166
00:16:02,440 --> 00:16:09,560
in this sense with large companies that have infinite resources between codes for this end

167
00:16:09,560 --> 00:16:16,840
user applications while having revenue from a different source same as us. Our company AG

168
00:16:16,840 --> 00:16:23,800
projects our source of revenue is maintenance that we do for telecom operators we also develop the

169
00:16:23,800 --> 00:16:29,880
server side for all these things that in the case of SIP you have a client but there is a server

170
00:16:29,880 --> 00:16:35,080
as well and these servers is a product that companies are ready to pay for and they have a

171
00:16:35,080 --> 00:16:40,760
budget for it and they have a maintenance budget for it and this is how our company was able to

172
00:16:40,760 --> 00:16:50,440
survive over time in order to build this for the end user free SIP clients like Blink and Sylk

173
00:16:50,600 --> 00:16:56,680
Again that development effort is huge because it's very complicated most of the internet

174
00:16:56,680 --> 00:17:02,680
is involved when you build such a client there are so many subparts and moving parts under

175
00:17:02,680 --> 00:17:08,920
the hood that you really have to be very knowledgeable and need to have people with

176
00:17:08,920 --> 00:17:14,680
the right experience to make it work as opposed to a simple application doing one little task

177
00:17:15,560 --> 00:17:21,480
If you look in under the hood of a complex application that does real-time communication

178
00:17:21,480 --> 00:17:28,120
its complex. So these people are hard to find and of course not cheap to build in the end such a

179
00:17:28,120 --> 00:17:34,920
software. We could do it because we had alternative sources of income and even so with those was not

180
00:17:34,920 --> 00:17:41,080
enough. Meaning that the effort is so high that without the help of NLnet and others alike, it

181
00:17:41,080 --> 00:17:48,040
is not possible to maintain and develop further a software that even after 14 years still works

182
00:17:48,040 --> 00:17:54,600
properly and it's up to date and still downloaded and used and validated by anyone who tries to use

183
00:17:54,600 --> 00:18:00,360
it has only good things to say about it. So this is a long-term effort that we're able with the help

184
00:18:00,360 --> 00:18:09,000
of NLnet in principle because they were the main sponsors over the years. And then my company during

185
00:18:09,000 --> 00:18:15,880
the idle financing was carrying forwards and continuing to maintain things over time so that

186
00:18:19,080 --> 00:18:24,280
things keep working and it's not just a dead project in a trench somewhere like many projects

187
00:18:24,280 --> 00:18:29,480
end up. After a short period of funding they just disappear and they just vanish because of lack of

188
00:18:29,480 --> 00:18:36,040
adoption or lack of support or lack of passion of somebody who maintains them. We did this long-term

189
00:18:36,040 --> 00:18:45,560
support and I'm proud of this personally. When I traveled back to my first office - it was last week -

190
00:18:46,200 --> 00:18:53,480
I found an old laptop, Del Inspiron laptop running Windows somewhere in a drawer. I picked it up and

191
00:18:53,480 --> 00:19:01,240
i tried to use it. It was from 2011 still worked I started it but I could not do anything with it. I

192
00:19:01,240 --> 00:19:07,080
could not read my emails, could not browse the internet. Everything was incompatible but I

193
00:19:07,080 --> 00:19:13,480
started Blink and I called my mother and it still worked. So this tells you something about the tools

194
00:19:13,480 --> 00:19:19,480
that IETF standardized and the way we do things that they are, how do you say, not timeless but

195
00:19:19,480 --> 00:19:28,360
resist long time. And this is one of the things that I'm very proud of that after so many years

196
00:19:28,360 --> 00:19:35,560
it still serves its purpose. I think this also taps into another question that I would have, 

197
00:19:37,160 --> 00:19:44,120
and this is not only related to the VoIP world but also to email,

198
00:19:45,640 --> 00:19:57,080
how is it possible to be a sustainable software company with open source software in this

199
00:19:57,080 --> 00:20:04,520
world of Big Tech. Like you said there are those those Big Tech companies they have a

200
00:20:05,240 --> 00:20:14,920
whole lot of developers, they have a lot of money and it's up to you with, well, less developers, less

201
00:20:14,920 --> 00:20:27,080
money to draw the competition. So how do you make it sustainable? Is that only by 

202
00:20:27,080 --> 00:20:36,520
other revenue sources? Is that only possible with grants? Well in my case the

203
00:20:36,520 --> 00:20:44,040
grants were not enough. And my company income dedicated for this part was also not enough. So

204
00:20:44,040 --> 00:20:49,880
somehow in my case I can definitely say that without both I could not have, for instance, just

205
00:20:49,880 --> 00:20:57,000
with NLnet I would not have been able to pull it off. It's just too much cost and there are unforeseen

206
00:20:57,800 --> 00:21:03,480
you know R&D is such that it's hard to predict the outcome. And you have to explore things that

207
00:21:03,480 --> 00:21:09,480
may have never been done before. And then you don't know how long it takes. So you may make a plan but

208
00:21:09,480 --> 00:21:17,960
you just run into unknown and then if you don't have resources, you basically lose all the investment

209
00:21:17,960 --> 00:21:26,280
at that point. So it's very hard with just one source of income it's definitely a big

210
00:21:26,280 --> 00:21:34,040
risk. With two it's better. The more you have... We had even a third source of income. We put one of

211
00:21:34,120 --> 00:21:40,600
the apps in the Mac app store where we asked money for it. Though there was a free version available

212
00:21:40,600 --> 00:21:48,360
for those who bothered to download the software from dangerous places like an website.

213
00:21:51,080 --> 00:21:55,800
This is something that it's so strange people yeah prefer to download it from the Apple store

214
00:21:55,800 --> 00:22:01,000
because it's more safe and pay for it because they think it's better but it's the same software.

215
00:22:01,640 --> 00:22:08,600
That helped but not much as well so the self revenues were never large enough to matter in

216
00:22:08,600 --> 00:22:16,200
the great scheme of things. Were too minimal to matter. But still everything adds up. So in the end

217
00:22:16,200 --> 00:22:21,240
because there is this persistent usage and the number of downloads is constant or growing or

218
00:22:21,240 --> 00:22:29,080
people buy it or you see there is constant demand for it. So when you see that and if you spend time

219
00:22:29,080 --> 00:22:34,760
doing it, you'll maintain it because it's your baby, so you don't want to see it dead. But that's of

220
00:22:34,760 --> 00:22:40,200
course because also I like what I do. A lot of people just follow the funding available without

221
00:22:40,200 --> 00:22:46,520
caring much about the outcome. And of course this matters. If you like what you do and if you really

222
00:22:46,520 --> 00:22:53,000
would do it even without the funding, for instance. It's different if it's a hobby or it's... So it

223
00:22:53,000 --> 00:22:58,280
depends on the person and the developer whether they can complement all these restrictions

224
00:22:58,280 --> 00:23:06,920
that appear. What general advice would you have for other open source

225
00:23:07,720 --> 00:23:16,360
companies to build a sustainable model? Well, that's like claiming I know what I'm doing.

226
00:23:16,360 --> 00:23:21,320
Business-wise I never had any clues. So I just follow my instincts. I don't have business plans,

227
00:23:22,760 --> 00:23:27,080
loans and my company was always cash flow positive and nobody taught me

228
00:23:27,080 --> 00:23:32,920
how to do it so I don't know how to give advice to other people because I have no plan I just

229
00:23:32,920 --> 00:23:39,080
use my instincts my instinct seems to be working and then that's very hard to answer questions.

230
00:23:41,480 --> 00:23:44,280
Well yeah, well, that's also honest. I mean

231
00:23:46,440 --> 00:23:55,000
yeah, maybe your instincts are good. Well they are as it as it turns out. But do understand correctly

232
00:23:55,000 --> 00:24:02,600
that the fact that Sylk is still alive is basically because you are so dedicated to it,

233
00:24:02,600 --> 00:24:12,040
that you will just continue to keep it... Yes, as long as I can I will. However, a

234
00:24:12,040 --> 00:24:17,560
little observation, this is actually Blink you should ask about because that's the old one. Let's

235
00:24:17,560 --> 00:24:24,760
say the old, the first one. Which is the pure SIP client. Sylk is very fresh, new and up to date with

236
00:24:24,760 --> 00:24:31,080
everything. Mobile and everything. So far from being over with it, it's like at the beginning of... 

237
00:24:31,880 --> 00:24:39,400
it's still in its infancy in terms of usage and the growth and adoption. So the real issue

238
00:24:39,400 --> 00:24:44,680
is Blink because some people may say that it's an obsolete piece of software which I totally disagree

239
00:24:44,680 --> 00:24:51,080
with. Just because, you know, it's like the cars today. They all suck. You want a nice looking car,

240
00:24:51,080 --> 00:24:58,280
that feels like a car you look for an older car that was a car. Today the

241
00:24:59,080 --> 00:25:03,480
principles of building cars have changed to the point of building an iPad and putting wheels

242
00:25:03,480 --> 00:25:11,080
around it. But that's not the car. That's something else. Same with everything. Beauty matters and

243
00:25:11,080 --> 00:25:18,600
classic things properly done never die. So we think Blink is the same and we try to maintain

244
00:25:18,600 --> 00:25:24,840
it very pure in sense of coding and the libraries and everything we use. It's still a

245
00:25:24,840 --> 00:25:31,080
a piece of software that made sense in 2009 and makes sense today if you start inspecting the code.

246
00:25:31,720 --> 00:25:36,280
So the code is written in such a way that is timeless. And as long as the

247
00:25:38,120 --> 00:25:43,800
support for the operating systems do not vanish, meaning in the end is based upon a programming

248
00:25:43,800 --> 00:25:48,440
language that of course may become obsolete over time, there's nothing you can do about that one for

249
00:25:48,440 --> 00:25:56,600
instance. But still the building blocks are pretty well alive and kicking. So there is no way that

250
00:25:56,600 --> 00:26:04,760
there is any obstacle in the near-term in the sense of survivability. And then the features

251
00:26:04,760 --> 00:26:11,480
matter. You try to be on par with the big boys which you can't because simply lack of funding

252
00:26:11,480 --> 00:26:17,160
and engineers and so forth. However things got so good that in the end a lot of features built

253
00:26:17,160 --> 00:26:23,160
nowadays are just not necessary and some companies just innovate for the sake of doing something

254
00:26:23,160 --> 00:26:28,120
though they shouldn't. So in this sense we don't have this problem. When we reach a level of

255
00:26:29,080 --> 00:26:36,920
maturity that we we like then we just keep it working without aiming to you know change it

256
00:26:36,920 --> 00:26:44,520
unnecessarily. So we leached we reached a level of maturity that requires conservation and

257
00:26:44,520 --> 00:26:50,360
incremental updates rather than completely retweaking, rechanging, changing everything. So

258
00:26:50,360 --> 00:26:54,600
we're very happy because it's comfortable to maintain something that makes sense, is beautiful,

259
00:26:54,600 --> 00:27:01,080
works fine you just add a few things. It's a pleasure, pleasureful work. And this is Blink. So

260
00:27:01,080 --> 00:27:07,000
it's a legacy piece of software that we keep it up to date in perfect shape. And thanks to

261
00:27:07,000 --> 00:27:12,600
NLnet we can bring it back to Windows. Because of lack of resources and lack of

262
00:27:12,600 --> 00:27:16,760
developers that over time we lost, there was a Windows version but at some point there was

263
00:27:16,760 --> 00:27:22,040
not anymore because we just couldn't do it. We didn't have the people and the time to maintain

264
00:27:22,040 --> 00:27:30,360
that part, that version. And we will be able to do this fall to bring Blink back to Windows

265
00:27:30,360 --> 00:27:36,600
because now it's available for Mac and Linux only. And with the help of NLnet we will be able to

266
00:27:36,600 --> 00:27:41,720
make it cross-platform truly again like it originally was for all three major desktop

267
00:27:41,720 --> 00:27:50,360
platforms. And again it's hard to know how to monetize this especially on Windows, I don't know

268
00:27:51,000 --> 00:27:56,680
if it's possible. And the goal also of NLnet is to allow for the software to be distributed purely

269
00:27:56,680 --> 00:28:01,720
in open source form and it's a condition of funding so I don't aim or plan to monetize

270
00:28:01,720 --> 00:28:11,000
necessarily Blink for Windows. So I don't know where it goes with this monetization of open

271
00:28:11,000 --> 00:28:17,800
source it's a difficult story. Again our story works because we have server side software that

272
00:28:17,800 --> 00:28:25,000
we monetize which indirectly funds the client side. And the client side helps as a marketing tool

273
00:28:25,000 --> 00:28:30,040
if - this could be an advice - whenever you leave, you give something for free, you can...

274
00:28:31,320 --> 00:28:37,560
In my company making free software that you publish it's like doing a marketing activity.

275
00:28:37,560 --> 00:28:41,720
People know you're knowledgeable in that area, so it's like a business card that

276
00:28:42,840 --> 00:28:47,960
people know and then when they need a solution for the server side they don't say: "but Adrian

277
00:28:47,960 --> 00:28:52,840
what do you know to do in this area?" They all know that I know because our software stands

278
00:28:53,480 --> 00:29:03,160
proof for the claims. How about the community behind Sylk is it only you as your company

279
00:29:03,160 --> 00:29:11,720
working on it? Is there an active community behind it? Right, there is a community but I

280
00:29:11,720 --> 00:29:17,880
wouldn't say developers. I mean the contributions can be seen on Github. We publish the software

281
00:29:17,880 --> 00:29:25,240
and all the patches. The authors that contributed over time. But the external contributions are

282
00:29:25,240 --> 00:29:31,800
a minimum to the level of patches and small changes. The heavy lifting still belongs to our company and

283
00:29:31,880 --> 00:29:35,960
developers dedicated full time for it. Because as I said, it's a complex task,

284
00:29:36,840 --> 00:29:43,000
takes time and it's... you need the proper individuals to even attempt to make a change to a software

285
00:29:43,000 --> 00:29:51,000
that is very complex. So in reality it's only our company developing it. And people using it may

286
00:29:51,000 --> 00:29:57,720
find, you know, little improvements, helping with translations and things that are marginal, in terms

287
00:29:57,720 --> 00:30:03,400
of there are no new functions that are very seldom happened or almost none that people added a

288
00:30:03,400 --> 00:30:09,640
feature but they did provide the fixes and things that we embedded back into the software

289
00:30:10,520 --> 00:30:17,240
with corrections and things that were discovered by end users. So the community really when I...

290
00:30:18,280 --> 00:30:25,320
To answer your question, the community in our case would be the SIP community. And the server side,

291
00:30:25,320 --> 00:30:31,240
which is related to OpenSIPS project, which is the server side of things. In order to make a SIP

292
00:30:31,240 --> 00:30:38,760
client or also webRTC client you need the correspondent server part. And without having

293
00:30:38,760 --> 00:30:46,120
that properly, how do you say, properly in place, you will fail to build a client because it's not just

294
00:30:46,120 --> 00:30:50,760
reading a spec and writing some software. You need a working server that would answer to all the

295
00:30:50,760 --> 00:30:58,200
requests your application is sending. And without having the server side ready and properly working,

296
00:30:58,200 --> 00:31:03,320
you cannot attempt to build a client. And actually this is the reason why our client works, because

297
00:31:03,320 --> 00:31:10,120
before even attempting to build a client, we had the server working properly. So everything we build

298
00:31:10,120 --> 00:31:15,720
in the client we could test accordingly and correct and fix. Whereas without having the

299
00:31:15,800 --> 00:31:20,680
counterpart, the server side, you cannot do it. Because it's impossible to just

300
00:31:21,720 --> 00:31:28,200
do it blindly with no actual proof it works. You need a server. Now in case this community

301
00:31:28,760 --> 00:31:35,320
of OpenSIPS goes back to the first open source project related to voice over IP.

302
00:31:36,120 --> 00:31:44,120
It was a project spawned by Fraunhofer Institute in Berlin. It was called SIP Express Router at the

303
00:31:44,120 --> 00:31:52,920
origin. And Fraunhofer Institute had director at the time who was looking for new ideas. That's

304
00:31:52,920 --> 00:32:00,040
back in time when Telecom was being contemplated being moved over the internet. And

305
00:32:00,040 --> 00:32:03,960
there were no standards… well, there were standards but there were no implementations, actually.

306
00:32:04,920 --> 00:32:10,840
Not in the open source domain. There were companies implementing standards but those were the old

307
00:32:10,840 --> 00:32:17,000
Telecom companies like Ericsson, Siemens and nothing in the open source domain and Fraunhofer

308
00:32:17,000 --> 00:32:25,640
Institute allocated a budget for developing the first SIP server in open source model. The director

309
00:32:27,080 --> 00:32:37,480
was a person of Romanian origins. And then he contacted his old friends in Bucharest and for

310
00:32:37,480 --> 00:32:45,960
this reason alone this project ended up full of Romanians. That's why I was also sort of around

311
00:32:45,960 --> 00:32:56,360
there perhaps. So Henry Sinnreich the mentor of this SIP Express Router project. Well actually, he helped

312
00:32:56,360 --> 00:33:03,480
guide the Fraunhofer ideas where to actually direct their development and R&D. So Henry was

313
00:33:03,480 --> 00:33:09,320
basically giving the clues. Then Fraunhofer spawned this software called SIP Express Router which, then,

314
00:33:10,520 --> 00:33:15,960
there were spin-offs, commercial ones, that took this software and made implementations

315
00:33:16,520 --> 00:33:22,600
in the market with it. Several companies did it. It became a success and it became the reference

316
00:33:22,600 --> 00:33:29,880
implementation of any operator that has SIP. Maybe they don't like to show it but I'm sure

317
00:33:29,880 --> 00:33:34,200
the developers and the people supporting can tell you: ‘Like yeah, of course, we still run

318
00:33:34,200 --> 00:33:41,720
SIP Express Router’, or a variant of it. Though they don't want to say too much about it. So the project

319
00:33:41,720 --> 00:33:48,680
emerged to be a success and this success helped also indirectly Blink. Because this community

320
00:33:49,320 --> 00:33:56,840
was present in many places at conferences, exhibitions and this ecosystem helped SIP

321
00:33:56,840 --> 00:34:03,720
protocol become obviously successful as well. And implicitly Blink and other software we

322
00:34:03,720 --> 00:34:10,440
created over time was present there and people knew us and then the community helped. That we

323
00:34:10,440 --> 00:34:18,040
all... all the companies present in this ecosystem benefited, including mine. So there was a

324
00:34:18,040 --> 00:34:25,800
complementary function in this ecosystem rather than competition. So Blink has no competition. I

325
00:34:25,800 --> 00:34:30,440
don't know any SIP client that can attempt to claim they have the same feature set. There are

326
00:34:30,440 --> 00:34:36,280
SIP clients but typically VoIP only or maybe they add another feature. But ours has all the features

327
00:34:36,280 --> 00:34:42,360
from the IETF working group. So it's the only one that has, or claims, to have the IETF working group

328
00:34:42,360 --> 00:34:52,520
stamp on. We did everything they standardized with SIP in the client. What besides working on

329
00:34:52,520 --> 00:34:59,560
the Windows client, what are the next steps for your project? Well, right now in the

330
00:34:59,560 --> 00:35:05,400
past couple of months, to attempt to actually do a Windows client, we had to

331
00:35:06,280 --> 00:35:11,720
bring the software up to date with the latest things that are today the dependencies on

332
00:35:11,720 --> 00:35:19,000
every computer. Over time they get modernized and you get to another version, namely QT framework

333
00:35:19,000 --> 00:35:24,520
which is the building block for Blink Qt which is the Linux and Windows version of Blink.

334
00:35:26,040 --> 00:35:32,200
This changed a lot so we had to basically redo a lot of the GUI elements. We had to

335
00:35:33,080 --> 00:35:38,840
basically port all the code that was originally written on Qt, I don't know, when we started this was

336
00:35:38,840 --> 00:35:45,960
Qt2 and now we are at Qt6. And there's a lot of work and heavy lifting to keep up with these changes.

337
00:35:45,960 --> 00:35:52,440
So, that was eating a lot of effort and we spent three months doing this. And now the software... just,

338
00:35:52,440 --> 00:35:57,720
we plan to make the first release, still on Linux. But we have all the

339
00:35:59,960 --> 00:36:08,120
code ported so it's compatible with Qt6. And then aim to have the building for Windows

340
00:36:08,120 --> 00:36:12,680
becomes possible. Because otherwise it wouldn't be. We could not just take the old code and make

341
00:36:12,680 --> 00:36:19,240
a build for windows. So plans after we do this, we're not there yet so we still

342
00:36:20,040 --> 00:36:26,680
work and we are still aiming for doing the Windows release. So this is the immediate things we're

343
00:36:27,560 --> 00:36:33,320
involved with. And I don't aim to dare where to go after that. Because I have my hands full already

344
00:36:34,280 --> 00:36:43,640
with this. So first step is to do the last release of Qt5 version which is still available on most

345
00:36:43,640 --> 00:36:52,360
Linux systems. But this year all new linux versions will have Qt6 as the default. So then immediately

346
00:36:52,360 --> 00:37:00,760
after we wrap with the old Qt5 legacy, we will almost be ready to go to the release for Qt6. Qt6 itself

347
00:37:00,760 --> 00:37:05,480
is not really ready to be released properly. It lacks skins for different operating systems, you

348
00:37:05,480 --> 00:37:12,680
have less flexibility in terms of, how do you say, it's a new thing but it's a future proof thing.

349
00:37:12,680 --> 00:37:19,880
Or well as future proof as you can make things for the next couple of years. So Qt6 is not really

350
00:37:19,880 --> 00:37:26,280
ready to be production ready but probably end of the year, next year it will be the de facto

351
00:37:27,000 --> 00:37:32,440
user interface widget block there is. So our software aims to be ready for it while we do

352
00:37:32,440 --> 00:37:40,200
this iteration. So when we do the Windows build, we will also have the Linux and the first incarnation

353
00:37:40,200 --> 00:37:49,320
of Mac based on this tool set. Because Blink had a different Mac source code, written differently. And

354
00:37:49,320 --> 00:37:55,080
with Qt now you can be really cross platform for all three operating systems. So you have the

355
00:37:55,080 --> 00:38:00,840
same code base on Linux, Mac and Windows. Whereas in the past we had Linux and Windows with the

356
00:38:00,840 --> 00:38:11,400
Blink Qt version and Mac had another version. And still has it which I'm afraid it will

357
00:38:11,400 --> 00:38:18,920
reach end of the line because of the conditions imposed by Apple. This is one of the things we

358
00:38:18,920 --> 00:38:24,200
are also battling with these ecosystems controlled by these large companies. They force you to

359
00:38:25,080 --> 00:38:30,600
make choices or to give up on traces based on their own interest. And suddenly you hit the wall,

360
00:38:30,600 --> 00:38:34,680
your software is not allowed in the App Store anymore because you cannot be compliant because 

361
00:38:34,680 --> 00:38:40,920
of reasons outside of your control. And I suspect the same will happen to the Blink Cocoa version

362
00:38:40,920 --> 00:38:47,560
at some point. Apple store will simply not allow it because the lack of support for Python language

363
00:38:47,560 --> 00:38:52,760
and the fact that the developers don't want to use it anymore in Mac, the Apple developers, I mean.

364
00:38:53,080 --> 00:39:00,680
They force you to basically dump your software because you can't fix the whole Apple libraries

365
00:39:00,680 --> 00:39:06,120
which are closed source that don't support various things anymore. And suddenly you update your

366
00:39:06,120 --> 00:39:11,960
computer and your software is dead. And there is no documentation at all and if you discover what the

367
00:39:11,960 --> 00:39:19,080
problem is, you have no support from Apple about it. They don't care and they force you to migrate

368
00:39:19,160 --> 00:39:24,920
in a different direction. Whereas the Qt seems to be going into the right direction, they have

369
00:39:24,920 --> 00:39:30,840
proper support on all platforms. So in the end probably Blink will be one version on all three

370
00:39:31,480 --> 00:39:36,200
operating systems. Now we still have two of them as I said. There is a Blink Qt version and

371
00:39:36,200 --> 00:39:43,080
a Blink Cocoa we call the native macOS incarnation of Blink. And Linux and Windows

372
00:39:43,080 --> 00:39:51,800
version based on a different widget set the Blink the Qt framework. Any considerations for

373
00:39:51,800 --> 00:39:59,240
iOS or Android clients or is that world already full of SIP clients?

374
00:39:59,880 --> 00:40:07,560
Well we did it, Sylk is a SIP client for both. Sylk as a client works on Apple and Android phones

375
00:40:07,560 --> 00:40:15,000
alike. So Sylk is not just a desktop solution but a mobile as well. And this is again where we

376
00:40:15,720 --> 00:40:21,400
stopped with Blink, in the sense of mobile developments, because it was a no-go because of

377
00:40:21,400 --> 00:40:28,760
the choice of the programming language namely Python. Python is not supported on

378
00:40:28,760 --> 00:40:35,000
mobiles. There is no Python anything on mobiles so basically it's not possible. It was not possible

379
00:40:35,080 --> 00:40:40,600
after starting the original desktop version to move towards mobile at the same software because

380
00:40:41,640 --> 00:40:47,640
it's impossible. And this is where, again, Sylk took a different approach and used the

381
00:40:48,680 --> 00:40:54,440
ools and the programming languages that were, how to say, properly supported in both sides,

382
00:40:56,440 --> 00:40:57,960
meaning Android and Apple.

383
00:40:58,440 --> 00:41:08,280
There is a React framework that is able to help developers to build once for both. So using this

384
00:41:08,280 --> 00:41:17,320
I built the Sylk Android version the Sylk iOS version and they still work fine. They are okay

385
00:41:17,320 --> 00:41:23,480
with their feature set which is less limited than well a WhatsApp or... But again it's with the amount

386
00:41:23,480 --> 00:41:31,640
of people and funding we did pretty well so far. It's an application that works and you can have

387
00:41:31,640 --> 00:41:37,960
features like replicated messages across multiple devices both desktop and mobile encryption end-to-

388
00:41:37,960 --> 00:41:45,080
end. A lot of good features that are hard to find in a similar project. And video conferencing from

389
00:41:45,080 --> 00:41:51,240
mobile and desktop with the limitations that are in the software. We're not perfect. But it's a

390
00:41:51,240 --> 00:41:56,360
perfectly workable solution for small groups. And you have the source code for everything and it's

391
00:41:56,360 --> 00:42:01,560
no surprise there. It's easy to inspect it, audit it, and you know what it does. As opposed to using

392
00:42:01,560 --> 00:42:07,560
other solutions that are black boxes and you don't know what happens to your data. Interesting and

393
00:42:09,960 --> 00:42:18,760
a recommendation, I think, for small groups to move to a private and and safe platform.

394
00:42:19,720 --> 00:42:28,120
Right. Yeah, which is probably a good bridge to a more general question. Because you have

395
00:42:28,920 --> 00:42:36,760
shown multiple ways in which there's actually a really beautiful standard and beautiful

396
00:42:36,760 --> 00:42:42,040
implementations but the the surrounding stuff of people wanting to commercialize it, or own it, or

397
00:42:42,840 --> 00:42:48,680
make it proprietary software, make it actually... take the good stuff and make it worse.

398
00:42:50,920 --> 00:42:57,480
Do you have any hopes, do you see any any ways, that this is going to take a different turn now?

399
00:42:57,480 --> 00:43:05,960
That either users or Big Tech are… I'm not sure. I'm a bit pessimistic because

400
00:43:06,680 --> 00:43:13,160
one of the issues we see is that a few players can decide overnight to change direction in ways

401
00:43:13,160 --> 00:43:20,040
that you you cannot do from your level. So we're all using this similar platforms that have similar

402
00:43:21,080 --> 00:43:29,640
interests. These cloud providers namely Google, Apple, all these people that we rely on them for

403
00:43:29,640 --> 00:43:37,640
our computers to run at all nowadays. They force upon us various changes that at the developer

404
00:43:37,640 --> 00:43:44,520
level you cannot change them. You have to go one, two levels up where policies are being made and

405
00:43:45,320 --> 00:43:51,960
carefully analyzed to see if it makes sense or not to do things that are being imposed on many of us

406
00:43:52,920 --> 00:44:01,960
One of the issues that I consider harmful is the misuse of secure identities today. We all claim

407
00:44:01,960 --> 00:44:07,080
we want that and there is a need for encryption but after all is said and done we end up with…

408
00:44:07,800 --> 00:44:14,120
the outcome is that for the end user suddenly it has, how do you say, is being just tracked all over

409
00:44:14,120 --> 00:44:21,240
the place. And while encryption is a good thing a lot of things don't have…. shouldn't even be encrypted.

410
00:44:22,040 --> 00:44:26,840
If I want to broadcast news, I don't need to have an encrypted session to, I don't know, a

411
00:44:26,840 --> 00:44:34,520
newspaper where they just, I don't know, just share the same information. And by imposing these TLS

412
00:44:34,520 --> 00:44:41,080
certificates everywhere, it becomes a nuisance to surf the web. All these certificates introduce a

413
00:44:41,080 --> 00:44:46,760
level of, they say security, but in the end, for the end user, ends up not being able to surf websites

414
00:44:46,760 --> 00:44:53,240
which are too old, they don't have the right certificate, and all these pop-ups that show up

415
00:44:53,240 --> 00:44:59,640
on your computer with things that nobody understands. I have to, you know, it's just... they're meant to

416
00:44:59,640 --> 00:45:05,960
scare you, so you just always go for their choice, to do their stuff, accept their conditions.

417
00:45:06,920 --> 00:45:12,760
And the people are slaves to this technology that is being controlled by few large players. And in

418
00:45:12,760 --> 00:45:21,080
this sense I don't know how people at my level can change it. We did make the internet run but

419
00:45:21,080 --> 00:45:27,400
now, once it runs, it's being taken over and hijacked by interests that are very high. And I don't know

420
00:45:28,200 --> 00:45:33,640
if we have the stamina to go there and fight. Because it's, I don't know, it's a pretty big

421
00:45:33,640 --> 00:45:42,040
battle and at my level I can't fight it because of lack of resources. But maybe to to add a little

422
00:45:42,040 --> 00:45:49,000
glimpse of hope in there. I do get the impression that people are getting more aware of the problems

423
00:45:49,000 --> 00:45:55,960
that it brings. And people in general and also governments and that there is already for a while

424
00:45:55,960 --> 00:46:03,080
people are pushing back... There is of course people look for alternatives as far as they can. And they

425
00:46:03,080 --> 00:46:07,720
do look and they do find alternatives. And we are such an alternative when it comes to real-time

426
00:46:07,800 --> 00:46:14,760
communications. But it's still confined to the developer type of persons. People who are

427
00:46:14,760 --> 00:46:20,040
aware because we marketed around SIP but nobody knows about SIP. People use a phone, they

428
00:46:20,040 --> 00:46:26,840
don't even know about technical things. Then if you try to make it something that is user-friendly

429
00:46:26,840 --> 00:46:34,040
- between quotes -suddenly it gets lost into the amalgam of other, similar solutions with small

430
00:46:34,040 --> 00:46:40,200
descriptions that nobody knows what they do and how. For that, then again, you make a client which

431
00:46:40,200 --> 00:46:44,600
doesn't... in the end looks the same like others. At least when you download it, or look at

432
00:46:44,600 --> 00:46:51,480
description. You can't tell what's the difference and then then they have better marketing. So you're

433
00:46:51,480 --> 00:46:58,200
lost there in trying to be normal, let's say, you get lost. Trying to become an expert in a certain

434
00:46:58,280 --> 00:47:05,320
area works but then your audience is confined to the experts in your ecosystem or, I don't know,

435
00:47:05,320 --> 00:47:11,640
people who are very tech-savvy and they say" 'Yes I want to have a SIP client where I understand my

436
00:47:11,640 --> 00:47:15,800
data, how it works.' But then again the other people you are calling, they don't care about this.

437
00:47:16,360 --> 00:47:21,240
They have WhatsApp and they don't want to change that even with telegram or other things. They just

438
00:47:21,240 --> 00:47:27,160
have WhatsApp and for them it’s the end. It's too complex already. So I don't know how to

439
00:47:28,360 --> 00:47:32,920
herd people in a certain direction when it comes to open source or open standards. It all makes

440
00:47:32,920 --> 00:47:37,800
sense but the adoption is forced by the default app installed on your phone when you buy it.

441
00:47:38,840 --> 00:47:44,520
So that's hard to fight, right? If you have a pre-installed application on a device you buy.

442
00:47:44,520 --> 00:47:50,040
And a billion people buy the same, then which application do you think people will use? Mine?

443
00:47:50,040 --> 00:47:56,920
No. So in this sense, I don't know, it's an uphill battle which is hard hard to win. Unless a third

444
00:47:56,920 --> 00:48:04,040
party comes with a different ecosystem, different mobile phone, different choices for software. But

445
00:48:04,040 --> 00:48:09,960
now you have none. You have Apple ecosystem and Google Android. And there is no other mobile

446
00:48:10,920 --> 00:48:17,400
phone provider even or operating system that is worthwhile noticing out there. So how can you change

447
00:48:17,400 --> 00:48:26,600
this. I can't change this. You can change it maybe. How would you do it. Well, again, to try to keep a

448
00:48:26,920 --> 00:48:33,720
bit of a glimpse of hope here. One thing I find very important is... The worst thing in the world

449
00:48:33,720 --> 00:48:38,760
is if you see a horrible problem and there is just you see no solution to it whatsoever. No

450
00:48:38,760 --> 00:48:44,680
no road to walk, no perspective that you can take… There are and I don't want to take the

451
00:48:44,680 --> 00:48:50,280
optimism... No, no, but that's where I'm going. It's like, people like you who are building

452
00:48:50,280 --> 00:48:57,960
this stuff, even indeed when it's really hard to get it out there because of this unequal playing

453
00:48:57,960 --> 00:49:03,880
field. For me, what gives me hope, is the fact that at least it's there, you

454
00:49:03,880 --> 00:49:09,880
know. If somebody comes to the realization this is no longer what I want, then there is a your

455
00:49:10,440 --> 00:49:16,040
projects but there's a whole host of stuff out there, that they can go to...

456
00:49:16,760 --> 00:49:22,200
Right, so it's good, as you mentioned, it is it's good to be optimistic.

457
00:49:22,920 --> 00:49:28,840
Don't take my pessimism... I'm looking specifically at my capabilities and my limitations. So I know

458
00:49:28,840 --> 00:49:35,800
how much I can do. And then of course, knowing how far you might want to go but you can't, you feel

459
00:49:35,800 --> 00:49:41,560
a bit frustrated. But of course, it's just a personal feeling it shouldn't be contagious.

460
00:49:41,560 --> 00:49:49,560
Well, I from our part, think that you are doing an awful lot actually to contribute to

461
00:49:50,920 --> 00:49:57,320
making this better, this problem… to addressing these issues. So maybe from your

462
00:49:57,320 --> 00:50:04,200
perspective, it's a bit negative but from my perspective you are doing an awful lot. So 

463
00:50:04,200 --> 00:50:10,360
thank you for that. Yeah well, it's a compliment, thank you very much. And maybe on that note,

464
00:50:12,520 --> 00:50:17,640
if the people who are listening to this podcast, if you.. What could they do to bring

465
00:50:18,280 --> 00:50:28,440
Blink and Sylk further. I think that the software stands for itself. And if you want

466
00:50:28,440 --> 00:50:34,680
to communicate over, let's say, you want something that works over time - look at email - you can have

467
00:50:34,680 --> 00:50:42,440
the same address for email as for Sylk. And because it's based on a standard you probably

468
00:50:42,440 --> 00:50:48,840
will still use it in 20 years. And I think this is, I’ll say, an advantage that many have missed,

469
00:50:49,640 --> 00:50:56,840
or nobody has emphasized enough. So but then again, it's what do you do if your phone

470
00:50:56,840 --> 00:51:02,200
number becomes your identifier. You know, you start the WhatsApp and it's your phone number.

471
00:51:02,920 --> 00:51:07,800
Where is your address, where is…? And then you're stuck with that and you can't even move your

472
00:51:07,800 --> 00:51:12,360
WhatsApp to another phone without great pain. And you're stuck in that ecosystem. With SIP you have

473
00:51:12,360 --> 00:51:18,760
the flexibility to have many devices. And you have this feature called parallel forking which is

474
00:51:19,400 --> 00:51:24,520
well SIP invented the fact that you can get called and accept the call in different places, at any

475
00:51:24,520 --> 00:51:32,440
moment. This was a SIP invention this parallel forking feature. This appeared later,

476
00:51:32,440 --> 00:51:39,880
much later, in WhatsApp on Telegram and other apps. But before you could not even do this. SIP

477
00:51:39,880 --> 00:51:47,080
protocol is the one that made it, actually, deployed it the first time. And they copied from it and there

478
00:51:47,080 --> 00:51:52,120
are a lot of things that SIP did right. And other did wrong, like with the video conferencing. So

479
00:51:52,600 --> 00:51:59,480
yeah, it's it's good to have alternatives. And then the problem is that all these alternatives

480
00:51:59,480 --> 00:52:07,160
had a lot... a short lifespan. And ours, we hope, it still has energy to go forward. And it did so far,

481
00:52:07,160 --> 00:52:13,560
for many years. And this is something that is important... I don't want to change my

482
00:52:13,560 --> 00:52:19,720
company programming language every six months. It's terrible. Same for communications, I don't want to

483
00:52:19,720 --> 00:52:24,840
look for another set of tools every… And SIP  is portable in the sense, if you have a...

484
00:52:25,640 --> 00:52:30,840
Blink, again, is not the only one you can put your address into another client. This alone gives you

485
00:52:30,840 --> 00:52:36,360
some sort of... how do you say, as an end user it gives you the freedom to use another

486
00:52:36,360 --> 00:52:42,360
software if you want. And you're not obliged to use... you have the coupling between the address

487
00:52:42,360 --> 00:52:46,920
and the client, it’s the same like your email address, yeah, you can put it in any email client on Windows

488
00:52:46,920 --> 00:52:53,400
or on your mobile it still works, right? This philosophy doesn't exist outside our ecosystem.

489
00:52:53,400 --> 00:52:59,400
What do you mean you take your address put it in another vendor’s SIP client and then you can call

490
00:52:59,400 --> 00:53:05,240
again with the same things? This is only in SIP possible and because of this ecosystem made it

491
00:53:05,240 --> 00:53:10,440
possible. The rest don't let you do that. If you have your WhatsApp identity you're not gonna

492
00:53:10,440 --> 00:53:15,880
run away with that anywhere else. They force you to just... all your contact list is stuck there.

493
00:53:16,520 --> 00:53:21,160
And of course, there are ways to import/export but there is no way to call a WhatsApp user from a SIP

494
00:53:21,160 --> 00:53:27,320
account. So they all close their islands and they have their user base... And it is actually the only

495
00:53:27,320 --> 00:53:33,640
way to cross this borders is by using SIP protocol, in reality. Anywhere there is any kind of

496
00:53:33,640 --> 00:53:40,760
interconnection between whatever islands, closed islands of... they may even use SIP internally or

497
00:53:40,920 --> 00:53:46,520
not, doesn't matter, the border is always SIP. To get the audio call or a video call carried across.

498
00:53:47,400 --> 00:53:53,960
So again, SIP is omnipresent, a good choice, and is being adopted quite a lot.

499
00:53:54,520 --> 00:54:02,120
But it's not advertised as such. So SIP it is a winner in this battlefield. But it's an invisible

500
00:54:02,120 --> 00:54:08,440
player, nobody knows about it. But it is there, as a building block, as a client, as a server,

501
00:54:09,000 --> 00:54:17,560
as an interconnect path between two islands of different solutions that need to interoperate

502
00:54:17,560 --> 00:54:26,760
with each other. So in this sense SIP did win the battle. But it is not deployed as such inside

503
00:54:27,720 --> 00:54:33,960
WhatsApp. Actually Apple themselves used SIP.  I don't know today anymore but they had a SIP

504
00:54:33,960 --> 00:54:39,320
client. They had face time running SIP. When the program crashed, you got a SIP dump out of it.

505
00:54:40,200 --> 00:54:45,960
They didn't say they used SIP until people saw the softer crashing. But you see, SIP is adopted.

506
00:54:45,960 --> 00:54:53,320
But then it's obfuscated by whomever is using it,tries to cover and hide it. So in this sense

507
00:54:53,320 --> 00:54:58,360
SIP is the winner. People try to say that we don't use it but we know…

508
00:54:58,360 --> 00:55:10,920
Well thank you very much for for unhiding SIP and showing it to us and giving it a... shining a little bit of light on it. Because it obviously deserves that.

509
00:55:12,360 --> 00:55:19,960
Ronny, do you have any more questions? Well I think not, technically, because this was

510
00:55:20,040 --> 00:55:27,800
a masterclass in SIP maybe. Yeah very interesting. 

511
00:55:28,600 --> 00:55:36,520
You also said multiple times that you were helped by the NGI funding

512
00:55:37,160 --> 00:55:43,480
and that you could not do your work without the NGI funding. So yeah, we had a question there but

513
00:55:44,200 --> 00:55:51,640
you answered that one already. But maybe, how would... what advice would you have for

514
00:55:51,640 --> 00:56:00,840
other people who are considering applying for NGI funding? Well advice, I don't have advice, but the

515
00:56:00,840 --> 00:56:07,000
reality is that a lot of funding is lost because, how do you say, no follow up and - but whose fault

516
00:56:07,000 --> 00:56:11,880
is it. I mean you may have a good idea but how to monetize it, how to survive after initial

517
00:56:11,880 --> 00:56:19,080
implementation. It's a problem for everyone. Advice... I remember starting my business. I was in amsterdam

518
00:56:19,080 --> 00:56:27,720
with the KvK, the chamber of commerce, and there was this seminar about startups. And there was this

519
00:56:27,720 --> 00:56:34,280
lady presenting and said like: 'You are all starting entrepreneurs' - and there are like probably a hundred

520
00:56:34,280 --> 00:56:41,240
of us there - 'do you know how many of you will still be doing what you came here for, and your

521
00:56:41,400 --> 00:56:46,920
business. And over one year time, how many of you will be still here?’ Or, no it was three years time.

522
00:56:46,920 --> 00:56:53,720
And everyone said like, well, maybe 50 percent. In the end the lady said: ‘Only three of you will still

523
00:56:53,720 --> 00:57:00,440
be doing what you aimed to do that you are here for today.’ Three out of a hundred and everyone booed

524
00:57:00,440 --> 00:57:07,880
and everyone said: ‘No way, our ideas are the best and we will be successful.’ And you need that to

525
00:57:07,960 --> 00:57:15,080
start. But then the reality is said a lot of things just don't have a future. They start but

526
00:57:15,080 --> 00:57:24,760
they don't go too far. And then advice? How can you give advice in this direction? It's

527
00:57:25,640 --> 00:57:31,480
a roller coaster. You have to observe… You have, first of all, to not just do things in a

528
00:57:31,480 --> 00:57:39,080
vacuum. You have to find an ecosystem, where, how do you say, you're not alone. And to be

529
00:57:39,080 --> 00:57:43,560
observed means there are others like you. So you cannot just be alone with the best solution in a

530
00:57:43,560 --> 00:57:50,840
place where nobody else is.  Because you'll not get noticed or… So whatever you do, try to find a

531
00:57:50,840 --> 00:57:55,960
place where you have more people like you. And go there and try to see if there is a place for what

532
00:57:55,960 --> 00:58:01,960
your idea is trying to achieve.  So don't be alone, go find your community. Find the people

533
00:58:02,760 --> 00:58:11,480
that are related to your field and try to get in there early. And especially if it's a growing

534
00:58:11,480 --> 00:58:17,320
trend it's good to be there early. And then even if you're late it's still not bad. But try to find

535
00:58:17,320 --> 00:58:24,200
a community where your solution fits. And try to start from there. Because funding alone will

536
00:58:24,200 --> 00:58:34,760
not solve your problems, long-term. Yeah, well thank you. That's good advice. Okay well, from

537
00:58:34,760 --> 00:58:40,600
our side I would say I'd like to say: thank you very much for this conversation. Thank you

538
00:58:40,600 --> 00:58:44,040
for inviting me to this session. 

539
00:58:54,200 --> 00:58:54,920
end