Connect with us

FACEBOOK

UX Best Practices: Using Emojis, Stickers and Gifs with Bots for Messenger

Published

on

Emojis, GIFs and stickers are commonly used in person to person communication, though they have slightly different uses and meanings. Here’s what we know about the use of these forms of communication between people:
  • Of the three types, emojis are more serious/formal in the information they convey, followed by stickers and GIFs.
  • When compared to emojis, people typically send stickers and GIFs to close friends and family members
  • GIFs are considered “funny” and “expressive.” Compared to emojis and stickers, GIFs are more expressive and can express emotions more precisely.
  • A single emoji can be sent as a form of expression, but are also often sent as a sequence of emojis (e.g. 🎉🎉🎉 to emphasize the sentiment or 🙄😜😂 to express more complex emotions).
  • Manage User Expectations: Be direct about what your bot can / can’t do. If you don’t want to provide support for gifs, images, etc. the best thing to do is to be clear with users that your bot doesn’t understand these. You can be direct and formal (“I’m sorry, I don’t know how to interpret images, gifs and stickers.”) or, based on your bot’s personality, you can respond with something more in character (“Sorry, I don’t have eyes! I can’t see images or gifs you send me.” or “My creators didn’t give me the capability to understand gifs and images. Can you help me by typing your response instead?”).
  • Train by Example: Just like with all bot outputs, if a bot says something to a user, then the expectation is that the bot will also understand the thing that it said. If you use GIFs, emojis and stickers in your responses to users, they are more likely to send them back to your bot, so be sure and provide support for bi-directional conversation based on the components your bot uses in conversation.
  • Understand sentiment: If possible, try to understand the sentiment behind the emoji that you’re responding to. Understanding sentiment is much harder for stickers and GIFs, but because emojis were standardized by the Unicode Consortium in 2010 as part of Unicode 6.0, there are tools out there that you can leverage for emojis. EmojiNet will provide keywords and similar emojis to the one being searched for, or can also search emojis based on a particular sentiment. There are also some sentiment analysis packages out there that can be used. Here are just a few: Github list of emoji rated for valence in JSON, Emoji > Sentiment library from NPM, Github emoji analysis
  • Provide meaningful output: For the most common emojis, you can hard code responses directly if you can’t integrate sentiment analysis. At the bare minimum, it is important to try and respond to the most common emojis and stickers from users with meaningful outputs (i.e. have a response and make sure it’s not a variant of “I don’t understand”). One of the most commonly used emoji/sticker in Facebook Messenger is the thumbs up sticker. You can consider this an acknowledgement from the user, so consider experimenting with responding as you would if the user said “ok” or “thanks” to your bot by using the thumbs up instead. Here’s an article that provides a list of the most common emojis from users in 2017 for Messenger.
  • Be cautious about mismatches in the bot response with the user sentiment: Don’t respond to a negative emoji, sticker, etc. with a positive response. It makes the user feel like they’re not being heard when providing negative feedback and it loses credibility. Additionally, there are very few circumstances in which ignoring user input is a good idea, so try not to ignore these inputs just because they’re different. Example of an interaction to avoid:

There are some significant advantages to using emojis, GIFs and stickers in your responses. It can be a great way to incorporate a bot personality that embodies your brand. It can also make the conversation more lighthearted, or more casual, which might be appropriate in some contexts. For example, recent research shows that adding emojis to engage users might work out well, but asking them questions about their physical activity might be better received without the use of emojis. However, if used, keep in mind that emojis do not look the same on every platform and they might have ambiguous meanings to users based on their cultural background and language.

Emojis come as text messages to your callback. Here are a couple of examples:

{ 
“sender”: { 
“id”: “<PSID>” 
},
“recipient”: { 
“id”: “<PAGE_ID>”
},
“timestamp”: 1458692752478,
“message”: { 
“mid”:”1B_UBJYvePYnQAOMIZLZJLC5640S_K72IRsYiZ2aIIz
FcDlDnKZPOllmCesLRzO_2YvwnX0VE2iVctjh-OmXVg”,
“seq”: 59333,
“text”: “:)”
} 
}{ 
"sender": { 
"id": "<PSID>" 
},
"recipient": { 
"id": "<PAGE_ID>"
},
"timestamp": 1544141335159,
"message": { 
"mid":"zFYmb2mX3vOeZzo62ZnRCbC5640S_K72IRsYiZ2aIIyvZGGRoG
RRevhV3qZwjiCA9z82sNv2ri-slKgIqLshFw",
"seq": 59382,
"text": "🙂"
} 
}

In both of the above cases, the user sees the same message sent to your bot, as Messenger will convert 🙂 to 🙂 for the user on screen. But because in one case the user typed “:)” and in another they selected the emoji from the emoji menu, your callback will receive slightly different input. Keep this in mind as you’re implementing your solution, so that your bot’s behavior is consistent across these two inputs, which are considered identical by users.

Images and GIFs have the exact same structure in the callback message your webhook receives. The developer documentation has a good example here.

Stickers have a slightly different format in the callback message. Here is an example of the thumbs up sticker that can be found on the bottom right corner of the Messenger user input bar.

{ 
"sender": { 
"id": "<PSID>" 
},
"recipient": { 
"id": "<PAGE_ID>"
},
"timestamp": 1544141768928,
"message": { 
"mid":"ZElljPP9ttpinKrBWi_9F7C5640S_K72IRsYiZ2aI
IxCiWwEDgmwqsDfPnRJeEaCT9YDHbK7rLtEJlJjvq0DWg",
"seq": 59406,
"sticker_id": 369239263222822,
"attachments": [ 
{
"type": "image",
"payload": {
"url": "https://scontent.xx.fbcdn.net/v/t39.1997- 6/39178562_1505197616293642_5411344281094848512_n.png?_nc_cat=1&_nc_log=1&_nc_oc=AWOixYXgN71O7TwlRODnEhedCoenAU8uS2YslmfhKRJlZxcSTLO_tXRQfH1lGtJtymQ37EKdtAY0XQ&_nc_ad=z-m&_nc_cid=0&_nc_ht=scontent.xx&oh=6156b594b1791351766973d936bebff1&oe=5C684675",
"sticker_id": 369239263222822
} 
} 
] 
}
}

In the example above, each sticker has a unique sticker_id. However, there’s no description of what the sticker means, or any way to look it up. The attachment is similar to the ones for images and GIFs and will give you the URL of the sticker, but will also contain the sticker_id again. During your design phase, you can choose not to support stickers at all (since they’re easily identifiable), or alternately you could choose to only support the most common stickers that users might send, and store your own mapping between the sticker and the sticker_id by trying the stickers out in your bot.

Developing a bot that handles emojis, GIFs, and stickers can be challenging, but when development is done and the bot is live, the experience between your bot <> audience can become colorful and imaginative. Have fun experimenting with different responses. Take the time to understand your audience and the way they want and love to interact. By using that information to build out your bot in a unique way, you will quickly find a recipe for success.

FACEBOOK

Youth apologises to parents on Facebook for ’embarrassing them’, hangs himself to death

Published

on

Youth apologises to parents on <b>Facebook</b> for 'embarrassing them', hangs himself to death thumbnail

The deceased was identified as Sumit Pardhe (Representative Image).

The deceased was identified as Sumit Pardhe (Representative Image).&nbsp | &nbspPhoto Credit:&nbspiStock Images

Key Highlights

  • A 24-year-old youth in Aurangabad allegedly hanged himself to death on Friday
  • The youth took the drastic after apologising to his parents on a Facebook Live

Aurangabad: A 24-year-old youth from Aurangabad, Maharashtra, allegedly ended his own life after apologising to his parents for “embarrassing them”. The youth went live on social media platform Facebook before taking the drastic step and apologised to his parents. 

The deceased was identified as Sumit Pardhe. Pardhe was found hanging from a tree in Paradh, Jalna on Friday morning. He was a resident of Hatti, Sillod tehsil of Aurangabad. 

‘The family is in shock and they are not in a position to speak’

Abhijit More, Paradh police station inspector said that the circumstances that prompted Pardhe to take the drastic step have not been ascertained yet. He added, “The family is in shock and they are not in a position to speak. “

The youth had gone to stay at his aunt’s home. On Friday morning, he left the house to go to a neighbouring farm where he allegedly hanged himself to death. Some of the locals saw the body and informed the police. The youth was taken to a nearby hospital where he was declared brought dead, The Times of India reported. 

Youth apologised for going against parents’ wishes 

The youth had completed his masters in science and used to play volleyball. During the Facebook Live session, the youth apologised to his parents for embarrassing them. He said that his parents had to apologise publicly because of him. The youth also said that his decision of going against his parents’ wishes caused all the problems for his family. 

Reportedly, the youth was disturbed over an incident that took place around three days before he took the extreme step. Efforts are underway to unearth the details of the incident. A case of accidental death was registered by the police. 


 

Read More

Continue Reading

FACEBOOK

Israel, Arabs and Jews: Was Facebook objective? – Analysis

Published

on

Last week, readers contacted The Jerusalem Post to suggest that we investigate claims that Facebook and Instagram were maliciously biasing the social media war against Israel, guided by powerful figures inside the company.

According to the claim, people pressing “report post” on blatantly antisemitic or anti-Israel content, or posts with false information about the recent military campaign, were told that the post “doesn’t violate our community guidelines.”

Reporters investigated a particular Instagram employee, a Muslim woman who has posted several pro-Palestinian images on her personal Instagram account, who activists said is one of the people who decide what is and isn’t in line with the social media giant’s community guidelines. “If the heads of these companies support these views themselves, why is it even surprising that no one sees our side?” one Jewish activist asked.

After investigating the matter further and speaking with a number of Facebook executives, the Post concluded that the accusation wasn’t strong enough to pursue. But an article published last week in Buzzfeed News made a similar accusation- from the Arab side.

According to the article, “Facebook is losing trust among Arab users,” because during the ongoing Palestinian-Israeli conflict, “censorship – either perceived or documented – had made Arab and Muslim users skeptical of the platform.” The article went on to list the same claims the Jewish activists had made, that their posts were being censored while the other side’s were not, and that powerful people inside the Facebook organization were making deliberately biased calls about what meets the company’s community standards and what does not.

The article quoted heavily from The Jerusalem Post’s September 2020 profile of Jordana Cutler, Facebook’s Head of Policy for Israel and the Jewish Diaspora, who was named one of the year’s most influential Jews. The article saw proof of Facebook’s pro-Israel bias in Cutler’s statements like “My job is to represent Facebook to Israel, and represent Israel to Facebook.” Facebook’s former head of policy for the Middle East and North Africa region, Ashraf Zeitoon, was quoted as saying he was “shocked” after seeing that interview.

Zeitoon, who left Facebook in 2017, shouldn’t have been so shocked though. Facebook maintains public policy teams in every country it works in, tasked with interfacing between the needs of the social media company and the legal and diplomatic needs of the local government.

“Jordana’s role, and the role of our public policy team around the world, is to help make sure local governments, regulators and civil society understand Facebook’s policies, and that we at Facebook understand the context of the countries where we operate. Jordana is part of a global policy team, and to suggest that her role is any kind of conflict of interest is entirely inaccurate and inflammatory,” a Facebook spokesperson said.

Israel, like other countries, expects Facebook to remove content that violates local laws, even if it meets Facebook’s own criteria. On that matter, Israel’s intervention during the Guardian of the Walls military campaign was relatively limited. Data from the cyber department of Israel’s Attorney-General shows that from May 8-26, Israeli officials made 608 requests from Facebook to remove posts, with 54% accepted. On Instagram, there were 190 official requests for removal, with a 46% acceptance rate.

The number of Israelis reporting hate speech and incitement through the platform seemingly had a far greater impact. According to Buzzfeed News, Israel, with 5.8 million Facebook users, reported to Facebook 550,000 posts violating policies for violence and hate speech and 155,000 posts for terrorist content during one week of fighting. During the period, Israelis reported 10 times more terrorism violations and eight times more hate violations compared to Palestinian users, Buzzfeed said, citing a company employee.

Zeitoon, in a different interview given to CBS News, attributed that gap to Israel’s organizational superiority. “Israel has hacked the system and knows how to pressure Facebook to take stuff down,” he was quoted as saying. “Palestinians don’t have the capacity, experience and resources to report hate speech by Israeli citizens in Hebrew.”

Others, however, note another difference: Hamas is recognized by many governments as a terrorist organization, and Palestinians posted in far greater number than Israelis direct calls for violence, hate speech, and content glorifying terrorism. Ignoring that aspect of the “Palestinian voice” that those like Zeitoon say is being suppressed is irresponsible and dangerous, they claim.

Israel is justifiably quite concerned about the clear and present dangers posed by social media. Reports in the Hebrew press suggest that Prime Minister Benjamin Netanyahu even proposed blocking social media sites completely in Israel as the recent conflict began, in hopes of quelling incitement. Many have referred to the recent uptick in violence as the TikTok Intifada, a reference to the video-sharing social media network that is particularly popular among a younger demographic, and is widely seen as the source of some of the most intense incitement activity against Israel.

Facebook, as well as TikTok, categorically asserts that its automated content removal tools and human content moderators show no systemic bias toward any political cause or movement.

On that post by the Israeli activist mentioned above, Facebook Israel communications manager Maayan Sarig responded sharply. “We take criticism very seriously, but false claims against specific employees are not acceptable. Our policies are conducted globally in accordance with our community rules and there is no content that is independently approved or removed by individuals. So let’s try to avoid conspiracy theories.” That sort of statement is echoed throughout the company’s internal and external communications.

TikTok likewise has told the Post that “Safety is our top priority and we do not tolerate violence, hate speech or hateful behavior.”

It is not surprising that people on both sides of the conflict accuse social platforms of being biased against their cause. But, as is often the case online, the nuances easily get drowned out by strong emotions.

Read More

Continue Reading

FACEBOOK

Facebook & Instagram will now allow all users to hide their like counts

Published

on

By

<b>Facebook</b> & Instagram will now allow all users to hide their like counts thumbnail

facebook

Facebook and Instagram are giving more control to users over their content, feed and privacy. 

This week they announced new tools such as a Feed Filter Bar, Favourite Feed and Choose Who Can Comment, which aim to give people more ways to control what they see on their news feeds.

Facebook has been working on another new tool that allows users to filter offensive content from their DMS, and they have been testing hiding like counts over the past months. 

The hiding like counts tool is “beneficial for some and annoying to others”, says Facebook.

They added, “We’re giving you the option to hide like counts on all posts in your feed. You’ll also have the option to hide like counts on your own posts, so others can’t see how many likes your posts get. This way, if you like, you can focus on the photos and videos being shared, instead of how many likes posts get.”

According to Facebook, “changing the way people view like counts is a big shift.” 

(Image Credit: www.thoughtcatalog.com with an active link required)

Read More

Continue Reading

Trending